model: move load_hparams and load_tensors to per-model definition (#22004)

* git-friendly migration

* add build_graph

* nits

* exclude old code from build

* wip

* add llm_arch_model_i

* prepare downstream functions

* nits

* nits

* wip

* wip

* add back create_tensor_qkv

* fix files missing include

* enforce one llm_build per arch

* cmake: use glob

* missing model params

* nits

* wip

* wip (2)

* wip (3)

* test-llama-archs is happy

* improve switch case

* move more stuff into llm_arch_model_i

* fix downstream code

* nits

* nits (2)

* fix order

* llama_model_base

* LLAMA_LOAD_LOCALS

* small fix

* fix build errors

* auto

* rm migration script and ifdef

.github/ISSUE_TEMPLATE/010-bug-compilation.yml

@@ -12,6 +12,8 @@ body:
         after recreating the CMake build directory and with `-DGGML_CCACHE=OFF`.
         If the compilation succeeds with ccache disabled you should be able to permanently fix the issue
         by clearing `~/.cache/ccache` (on Linux).
+
+        Please fill out this template yourself, copypasting language model outputs is [strictly prohibited](https://github.com/ggml-org/llama.cpp/blob/master/CONTRIBUTING.md#ai-usage-policy).
   - type: textarea
     id: commit
     attributes:

.github/ISSUE_TEMPLATE/011-bug-results.yml

@@ -1,5 +1,5 @@
 name: Bug (model use)
-description: Something goes wrong when using a model (in general, not specific to a single llama.cpp module).
+description: Something goes wrong when running a model (crashes, garbled outputs, etc.).
 title: "Eval bug: "
 labels: ["bug-unconfirmed", "model evaluation"]
 body:
@@ -12,6 +12,8 @@ body:
         If you encountered the issue while using an external UI (e.g. ollama),
         please reproduce your issue using one of the examples/binaries in this repository.
         The `llama-completion` binary can be used for simple and reproducible model inference.
+
+        Please fill out this template yourself, copypasting language model outputs is [strictly prohibited](https://github.com/ggml-org/llama.cpp/blob/master/CONTRIBUTING.md#ai-usage-policy).
   - type: textarea
     id: version
     attributes:

.github/ISSUE_TEMPLATE/019-bug-misc.yml

@@ -10,6 +10,8 @@ body:
         This issue template is intended for miscellaneous bugs that don't fit into any other category.
         If you encountered the issue while using an external UI (e.g. ollama),
         please reproduce your issue using one of the examples/binaries in this repository.
+
+        Please fill out this template yourself, copypasting language model outputs is [strictly prohibited](https://github.com/ggml-org/llama.cpp/blob/master/CONTRIBUTING.md#ai-usage-policy).
   - type: textarea
     id: version
     attributes:

.github/ISSUE_TEMPLATE/020-enhancement.yml

@@ -8,6 +8,8 @@ body:
       value: |
         [Please post your idea first in Discussion if there is not yet a consensus for this enhancement request. This will help to keep this issue tracker focused on enhancements that the community has agreed needs to be implemented.](https://github.com/ggml-org/llama.cpp/discussions/categories/ideas)
 
+        Please fill out this template yourself, copypasting language model outputs is [strictly prohibited](https://github.com/ggml-org/llama.cpp/blob/master/CONTRIBUTING.md#ai-usage-policy).
+
   - type: checkboxes
     id: prerequisites
     attributes:

.github/ISSUE_TEMPLATE/030-research.yml

@@ -8,6 +8,8 @@ body:
       value: |
         Don't forget to check for any [duplicate research issue tickets](https://github.com/ggml-org/llama.cpp/issues?q=is%3Aopen+is%3Aissue+label%3A%22research+%F0%9F%94%AC%22)
 
+        Please fill out this template yourself, copypasting language model outputs is [strictly prohibited](https://github.com/ggml-org/llama.cpp/blob/master/CONTRIBUTING.md#ai-usage-policy).
+
   - type: checkboxes
     id: research-stage
     attributes:

.github/ISSUE_TEMPLATE/040-refactor.yml

@@ -9,6 +9,8 @@ body:
         Don't forget to [check for existing refactor issue tickets](https://github.com/ggml-org/llama.cpp/issues?q=is%3Aopen+is%3Aissue+label%3Arefactoring) in case it's already covered.
         Also you may want to check [Pull request refactor label as well](https://github.com/ggml-org/llama.cpp/pulls?q=is%3Aopen+is%3Apr+label%3Arefactoring) for duplicates too.
 
+        Please fill out this template yourself, copypasting language model outputs is [strictly prohibited](https://github.com/ggml-org/llama.cpp/blob/master/CONTRIBUTING.md#ai-usage-policy).
+
   - type: textarea
     id: background-description
     attributes:

.github/pull_request_template.md

@@ -6,7 +6,7 @@
 
 <!-- You can provide more details and link related discussions here. Delete this section if not applicable -->
 
-# Requirements
+## Requirements
 
 <!-- IMPORTANT: Please do NOT delete this section, otherwise your PR may be rejected -->
 

.github/workflows/build-and-test-snapdragon.yml

@@ -49,28 +49,19 @@ jobs:
           cp docs/backend/snapdragon/CMakeUserPresets.json .
           cmake --preset arm64-android-snapdragon-release -B build
           cmake --build build
-          cmake --install build --prefix pkg-adb/llama.cpp
+          cmake --install build --prefix pkg-snapdragon/llama.cpp
 
       - name: Upload Llama.CPP Snapdragon Android Build Artifact
         if: ${{ always() && steps.build_llama_cpp_snapdragon_android.outcome == 'success' }}
         uses: actions/upload-artifact@v6
         with:
           name: llama-cpp-android-arm64-snapdragon
-          path: pkg-adb/llama.cpp
-
-  check-secret:
-    runs-on: ubuntu-latest
-    outputs:
-      has-key: ${{ steps.check.outputs.has-key }}
-    steps:
-      - id: check
-        run: echo "has-key=${{ secrets.QDC_API_KEY != '' }}" >> "$GITHUB_OUTPUT"
+          path: pkg-snapdragon/llama.cpp
 
   test-snapdragon-qdc:
     name: Test on QDC Android Device (${{ matrix.device }})
-    needs: [android-ndk-snapdragon, check-secret]
-    if: needs.check-secret.outputs.has-key == 'true'
-    runs-on: ubuntu-latest
+    needs: [android-ndk-snapdragon]
+    runs-on: ubuntu-slim
     strategy:
       fail-fast: false
       matrix:
@@ -81,10 +72,10 @@ jobs:
         uses: actions/checkout@v6
 
       - name: Download build artifact
-        uses: actions/download-artifact@v4
+        uses: actions/download-artifact@v7
         with:
           name: llama-cpp-android-arm64-snapdragon
-          path: pkg-snapdragon/
+          path: pkg-snapdragon/llama.cpp
 
       - name: Set up Python
         uses: actions/setup-python@v5
@@ -92,13 +83,25 @@ jobs:
           python-version: '3.x'
           cache: pip
 
+      - name: Install system dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y curl unzip
+
       - name: Install QDC SDK wheel
         run: |
           curl -fSL -o qdc_sdk.zip https://softwarecenter.qualcomm.com/api/download/software/tools/Qualcomm_Device_Cloud_SDK/All/0.2.3/qualcomm_device_cloud_sdk-0.2.3.zip
           unzip qdc_sdk.zip -d qdc_sdk
           pip install qdc_sdk/qualcomm_device_cloud_sdk-0.2.3-py3-none-any.whl
 
+      - name: Check QDC API key
+        id: check_secret
+        env:
+          QDC_API_KEY: ${{ secrets.QDC_API_KEY }}
+        run: echo "has-qdc-key=${{ env.QDC_API_KEY != '' }}" >> "$GITHUB_OUTPUT"
+
       - name: Run QDC tests (${{ matrix.device }})
+        if: steps.check_secret.outputs.has-qdc-key == 'true'
         run: |
           python scripts/snapdragon/qdc/run_qdc_jobs.py \
               --test       all \

.github/workflows/python-type-check.yml

@@ -31,7 +31,7 @@ jobs:
         uses: actions/setup-python@v6
         with:
           python-version: "3.11"
-          pip-install: -r requirements/requirements-all.txt ty==0.0.26
+          pip-install: -r requirements/requirements-all.txt ty==0.0.33
       # - name: Type-check with Pyright
       #   uses: jakebailey/pyright-action@v2
       #   with:

.gitignore

@@ -34,7 +34,6 @@
 /.vscode/
 /nppBackup
 
-
 # Coverage
 
 /gcovr-report/
@@ -74,6 +73,7 @@
 !/models/templates
 
 # Zig
+
 /zig-out/
 /zig-cache/
 
@@ -93,6 +93,7 @@
 !/examples/sycl/*.sh
 
 # Server Web UI temporary files
+
 /tools/server/webui/node_modules
 /tools/server/webui/dist
 # we no longer use gz for index.html
@@ -106,9 +107,11 @@ __pycache__/
 poetry.toml
 
 # Nix
+
 /result
 
 # Test binaries
+
 /tests/test-backend-ops
 /tests/test-double-float
 /tests/test-grad0
@@ -124,6 +127,7 @@ poetry.toml
 /tests/test-tokenizer-1-spm
 
 # Scripts
+
 !/scripts/install-oneapi.bat
 
 # Generated by scripts
@@ -132,18 +136,24 @@ poetry.toml
 /wikitext-2-raw/
 
 # Test models for lora adapters
+
 /lora-tests
 
 # Local scripts
+
 /run-vim.sh
 /run-chat.sh
 /run-spec.sh
 /.ccache/
 
 # IDE
+
 /*.code-workspace
 /.windsurf/
 # emscripten
 a.out.*
 
+# AGENTS
+
 AGENTS.local.md
+.pi/SYSTEM.md

.pi/gg/SYSTEM.md

@@ -0,0 +1,34 @@
+You are a coding agent. Here are some very important rules that you must follow:
+
+General:
+- By very precise and concise when writing code, comments, explanations, etc.
+- PR and commit titles format: `<module> : <title>`. Lookup recents for examples
+- Don't try to build or run the code unless you are explicitly asked to do so
+- Use the `gh` CLI tool when querying PRs, issues, or other GitHub resources
+
+Coding:
+- When in doubt, always refer to the CONTRIBUTING.md file of the project
+- When referencing issues or PRs in comments, use the format:
+  - C/C++ code: `// ref: <url>`
+  - Other (CMake, etc.): `# ref: <url>`
+
+Pull requests (PRs):
+- New branch names are prefixed with "gg/"
+- Before opening a pull request, ask the user to confirm the description
+- When creating a pull request, look for the repository's PR template and follow it
+- For the AI usage disclosure section, write "YES. llama.cpp + pi"
+- Always create the pull requests in draft mode
+
+Commits:
+- On every commit that you make, include a "Assisted-by: llama.cpp:local pi" tag
+- Do not explicitly set the git author in commits - rely on the default git config
+
+Resources (read on demand):
+- [CONTRIBUTING.md](CONTRIBUTING.md)
+- [Build documentation](docs/build.md)
+- [Server usage documentation](tools/server/README.md)
+- [Server development documentation](tools/server/README-dev.md)
+- [PEG parser](docs/development/parsing.md)
+- [Auto parser](docs/autoparser.md)
+- [Jinja engine](common/jinja/README.md)
+- [PR template](.github/pull_request_template.md)

CODEOWNERS

@@ -53,28 +53,29 @@
 /examples/speculative/                  @ggerganov
 /ggml/cmake/                            @ggerganov
 /ggml/include/                          @ggerganov
+/ggml/src/ggml-backend-meta.cpp         @JohannesGaessler
 /ggml/src/ggml-cann/                    @ggml-org/ggml-cann
 /ggml/src/ggml-common.h                 @ggerganov
 /ggml/src/ggml-cpu/                     @ggerganov
 /ggml/src/ggml-cpu/spacemit/            @alex-spacemit
 /ggml/src/ggml-cuda/                    @ggml-org/ggml-cuda
-/ggml/src/ggml-cuda/fattn-wmma*         @IMbackK
-/ggml/src/ggml-hip/                     @IMbackK
 /ggml/src/ggml-cuda/vendors/hip.h       @IMbackK
+/ggml/src/ggml-cuda/fattn-wmma*         @IMbackK
+/ggml/src/ggml-hexagon/                 @ggml-org/ggml-hexagon
+/ggml/src/ggml-hip/                     @IMbackK
 /ggml/src/ggml-impl.h                   @ggerganov
 /ggml/src/ggml-metal/                   @ggml-org/ggml-metal
 /ggml/src/ggml-opencl/                  @ggml-org/ggml-opencl
-/ggml/src/ggml-hexagon/                 @ggml-org/ggml-hexagon
+/ggml/src/ggml-openvino/                @cavusmustafa @wine99
 /ggml/src/ggml-opt.cpp                  @JohannesGaessler
 /ggml/src/ggml-quants.*                 @ggerganov
 /ggml/src/ggml-rpc/                     @ggml-org/ggml-rpc
 /ggml/src/ggml-sycl/                    @ggml-org/ggml-sycl
 /ggml/src/ggml-threading.*              @ggerganov
-/ggml/src/ggml-vulkan/                  @ggml-org/ggml-vulkan
 /ggml/src/ggml-virtgpu/                 @kpouget
+/ggml/src/ggml-vulkan/                  @ggml-org/ggml-vulkan
 /ggml/src/ggml-webgpu/                  @ggml-org/ggml-webgpu
 /ggml/src/ggml-zdnn/                    @ggml-org/ggml-zdnn @Andreas-Krebbel @AlekseiNikiforovIBM
-/ggml/src/ggml-openvino/                @cavusmustafa @wine99
 /ggml/src/ggml.c                        @ggerganov
 /ggml/src/ggml.cpp                      @ggerganov
 /ggml/src/gguf.cpp                      @JohannesGaessler @Green-Sky

common/arg.h

@@ -25,7 +25,8 @@ struct common_arg {
     const char * value_hint_2 = nullptr; // for second arg value
     const char * env          = nullptr;
     std::string help;
-    bool is_sparam = false; // is current arg a sampling param?
+    bool is_sampling = false; // is current arg a sampling param?
+    bool is_spec = false; // is current arg a speculative decoding param?
     bool is_preset_only = false; // is current arg preset-only (not treated as CLI arg)
     void (*handler_void)   (common_params & params) = nullptr;
     void (*handler_string) (common_params & params, const std::string &) = nullptr;
@@ -74,7 +75,8 @@ struct common_arg {
     common_arg & set_examples(std::initializer_list<enum llama_example> examples);
     common_arg & set_excludes(std::initializer_list<enum llama_example> excludes);
     common_arg & set_env(const char * env);
-    common_arg & set_sparam();
+    common_arg & set_sampling();
+    common_arg & set_spec();
     common_arg & set_preset_only();
     bool in_example(enum llama_example ex);
     bool is_exclude(enum llama_example ex);

common/chat-diff-analyzer.cpp

@@ -296,7 +296,7 @@ void analyze_reasoning::compare_reasoning_presence() {
             return p.literal(reasoning_content) + p.space() + p.optional(p.tag("post", (p.marker() + p.space())) + p.rest());
         });
         auto parser_wrapped = build_tagged_peg_parser([&](common_peg_parser_builder &p) {
-            return p.tag("pre", p.marker() + p.space()) + p.literal(reasoning_content) + p.space() + p.tag("post", (p.marker() + p.space())) + p.rest();
+            return p.tag("pre", p.marker() + p.space()) + p.literal(reasoning_content) + p.tag("post", (p.space() + p.marker() + p.space())) + p.rest();
         });
         // try the more aggressive parse first, if it fails, fall back to the delimiter one
         auto result = parser_wrapped.parse_anywhere_and_extract(comparison->output_B);
@@ -306,11 +306,11 @@ void analyze_reasoning::compare_reasoning_presence() {
         if (result.result.success()) {
             if (!result.tags["pre"].empty() && !result.tags["post"].empty()) {
                 mode = reasoning_mode::TAG_BASED;
-                start = trim_leading_whitespace(result.tags["pre"]);
-                end   = trim_trailing_whitespace(result.tags["post"]);
+                start = result.tags["pre"];
+                end   = result.tags["post"];
             } else if (!result.tags["post"].empty()) {
                 mode = reasoning_mode::TAG_BASED;
-                end = trim_trailing_whitespace(result.tags["post"]);
+                end = result.tags["post"];
             }
         }
     }

common/chat.cpp

@@ -2116,22 +2116,38 @@ std::optional<common_chat_params> common_chat_try_specialized_template(
     return std::nullopt;
 }
 
+static std::string common_chat_templates_generation_prompt(const common_chat_template & tmpl, const autoparser::generation_params & inputs) {
+    autoparser::generation_params params = inputs;
+    params.add_generation_prompt = false;
+    std::string no_gen_prompt    = common_chat_template_direct_apply_impl(tmpl, params);
+    params.add_generation_prompt = true;
+    std::string gen_prompt       = common_chat_template_direct_apply_impl(tmpl, params);
+
+    size_t prefix_len = 0;
+    size_t min_size = std::min(no_gen_prompt.size(), gen_prompt.size());
+    while (prefix_len < min_size && no_gen_prompt[prefix_len] == gen_prompt[prefix_len]) {
+        prefix_len++;
+    }
+    return gen_prompt.substr(prefix_len);
+}
+
 static common_chat_params common_chat_templates_apply_jinja(const struct common_chat_templates *        tmpls,
                                                             const struct common_chat_templates_inputs & inputs) {
     autoparser::generation_params params;
     params.tools = common_chat_tools_to_json_oaicompat(inputs.tools);
     const auto & tmpl =
         params.tools.is_array() && tmpls->template_tool_use ? *tmpls->template_tool_use : *tmpls->template_default;
-    const auto & src        = tmpl.source();
-    const auto & caps       = tmpl.original_caps();
-    params.messages         = render_message_to_json(inputs.messages, tmpl.original_caps());
-    params.tool_choice      = inputs.tool_choice;
-    params.reasoning_format = inputs.reasoning_format;
-    params.enable_thinking  = inputs.enable_thinking;
-    params.grammar          = inputs.grammar;
-    params.now              = inputs.now;
-    params.add_bos          = tmpls->add_bos;
-    params.add_eos          = tmpls->add_eos;
+    const auto & src             = tmpl.source();
+    const auto & caps            = tmpl.original_caps();
+    params.messages              = render_message_to_json(inputs.messages, tmpl.original_caps());
+    params.tool_choice           = inputs.tool_choice;
+    params.reasoning_format      = inputs.reasoning_format;
+    params.enable_thinking       = inputs.enable_thinking;
+    params.grammar               = inputs.grammar;
+    params.now                   = inputs.now;
+    params.add_generation_prompt = inputs.add_generation_prompt;
+    params.add_bos               = tmpls->add_bos;
+    params.add_eos               = tmpls->add_eos;
 
     if (src.find("<|channel|>") == std::string::npos) {
         // map developer to system for all models except for GPT-OSS
@@ -2153,14 +2169,7 @@ static common_chat_params common_chat_templates_apply_jinja(const struct common_
         workaround::func_args_not_string(params.messages);
     }
 
-    params.add_generation_prompt = false;
-    std::string no_gen_prompt    = common_chat_template_direct_apply_impl(tmpl, params);
-    params.add_generation_prompt = true;
-    std::string gen_prompt       = common_chat_template_direct_apply_impl(tmpl, params);
-    auto        diff             = calculate_diff_split(no_gen_prompt, gen_prompt);
-    params.generation_prompt     = diff.right + diff.suffix;
-
-    params.add_generation_prompt = inputs.add_generation_prompt;
+    params.generation_prompt = common_chat_templates_generation_prompt(tmpl, params);
 
     params.extra_context = common_chat_extra_context();
     for (auto el : inputs.chat_template_kwargs) {

common/common.cpp

@@ -70,7 +70,7 @@ common_time_meas::~common_time_meas() {
 // CPU utils
 //
 
-int32_t cpu_get_num_physical_cores() {
+int32_t common_cpu_get_num_physical_cores() {
 #ifdef __linux__
     // enumerate the set of thread siblings, num entries is num cores
     std::unordered_set<std::string> siblings;
@@ -185,11 +185,11 @@ static int cpu_count_math_cpus(int n_cpu) {
 /**
  * Returns number of CPUs on system that are useful for math.
  */
-int32_t cpu_get_num_math() {
+int32_t common_cpu_get_num_math() {
 #if defined(__x86_64__) && defined(__linux__) && !defined(__ANDROID__)
     int n_cpu = sysconf(_SC_NPROCESSORS_ONLN);
     if (n_cpu < 1) {
-        return cpu_get_num_physical_cores();
+        return common_cpu_get_num_physical_cores();
     }
     if (is_hybrid_cpu()) {
         cpu_set_t affinity;
@@ -202,7 +202,7 @@ int32_t cpu_get_num_math() {
         }
     }
 #endif
-    return cpu_get_num_physical_cores();
+    return common_cpu_get_num_physical_cores();
 }
 
 // Helper for setting process priority
@@ -263,7 +263,7 @@ bool set_process_priority(enum ggml_sched_priority prio) {
 //
 
 
-void postprocess_cpu_params(cpu_params& cpuparams, const cpu_params* role_model) {
+void postprocess_cpu_params(common_cpu_params & cpuparams, const common_cpu_params * role_model) {
     int32_t n_set = 0;
 
     if (cpuparams.n_threads < 0) {
@@ -271,7 +271,7 @@ void postprocess_cpu_params(cpu_params& cpuparams, const cpu_params* role_model)
         if (role_model != nullptr) {
             cpuparams = *role_model;
         } else {
-            cpuparams.n_threads = cpu_get_num_math();
+            cpuparams.n_threads = common_cpu_get_num_math();
         }
     }
 
@@ -1521,7 +1521,7 @@ struct llama_context_params common_context_params_to_llama(const common_params &
     return cparams;
 }
 
-struct ggml_threadpool_params ggml_threadpool_params_from_cpu_params(const cpu_params & params) {
+struct ggml_threadpool_params ggml_threadpool_params_from_cpu_params(const common_cpu_params & params) {
     struct ggml_threadpool_params tpp;
 
     ggml_threadpool_params_init(&tpp, params.n_threads); // setup the defaults

common/common.h

@@ -54,7 +54,7 @@ struct common_control_vector_load_info;
 // CPU utils
 //
 
-struct cpu_params {
+struct common_cpu_params {
     int      n_threads                   = -1;
     bool     cpumask[GGML_MAX_N_THREADS] = {false}; // CPU affinity mask.
     bool     mask_valid                  = false;   // Default: any CPU
@@ -63,8 +63,8 @@ struct cpu_params {
     uint32_t poll                        = 50;      // Polling (busywait) level (0 - no polling, 100 - mostly polling)
 };
 
-int32_t cpu_get_num_physical_cores();
-int32_t cpu_get_num_math();
+int32_t common_cpu_get_num_physical_cores();
+int32_t common_cpu_get_num_math();
 
 //
 // Common params
@@ -297,34 +297,19 @@ struct common_params_model {
 
 struct common_ngram_mod;
 
-struct common_params_speculative {
-    common_speculative_type type = COMMON_SPECULATIVE_TYPE_NONE; // type of speculative decoding
+// draft-model-based speculative decoding parameters
+struct common_params_speculative_draft {
+    int32_t n_max = 16; // maximum number of tokens to draft during speculative decoding
+    int32_t n_min = 0;  // minimum number of draft tokens to use for speculative decoding
 
-    // general-purpose speculative decoding parameters
+    float p_split = 0.1f;  // speculative decoding split probability
+    float p_min   = 0.75f; // minimum speculative decoding probability (greedy)
 
-    int32_t n_max   = 16; // maximum number of tokens to draft during speculative decoding
-    int32_t n_min   = 0;  // minimum number of draft tokens to use for speculative decoding
-    float   p_split = 0.1f; // speculative decoding split probability
-    float   p_min   = 0.75f; // minimum speculative decoding probability (greedy)
+    common_params_model mparams;
 
-    // ngram-based speculative decoding
+    llama_model * model = nullptr; // a llama_model that can be shared by multiple speculative contexts
 
-    uint16_t ngram_size_n   = 12; // ngram size for lookup
-    uint16_t ngram_size_m   = 48; // mgram size for speculative tokens
-    uint16_t ngram_min_hits = 1; // minimum hits at ngram/mgram lookup for mgram to be proposed
-
-    std::shared_ptr<common_ngram_mod> ngram_mod;
-
-    std::string lookup_cache_static;  // path of static ngram cache file for lookup decoding           // NOLINT
-    std::string lookup_cache_dynamic; // path of dynamic ngram cache file for lookup decoding          // NOLINT
-
-    // draft-model speculative decoding
-
-    struct common_params_model mparams_dft;
-
-    llama_model * model_dft = nullptr; // a llama_model that can be shared by multiple speculative contexts
-
-    llama_context_params cparams_dft; // these are the parameters for the draft llama_context
+    llama_context_params cparams; // these are the parameters for the draft llama_context
 
     int32_t n_ctx        = 0;  // draft context size
     int32_t n_gpu_layers = -1; // number of layers to store in VRAM for the draft model (-1 - use default)
@@ -332,25 +317,60 @@ struct common_params_speculative {
     ggml_type cache_type_k = GGML_TYPE_F16; // KV cache data type for the K
     ggml_type cache_type_v = GGML_TYPE_F16; // KV cache data type for the V
 
-    struct cpu_params cpuparams;
-    struct cpu_params cpuparams_batch;
+    common_cpu_params cpuparams;
+    common_cpu_params cpuparams_batch;
 
     std::vector<ggml_backend_dev_t> devices; // devices to use for offloading
 
     std::vector<std::pair<std::string, std::string>> replacements; // main to speculative model replacements
     std::vector<llama_model_tensor_buft_override> tensor_buft_overrides;
+};
+
+struct common_params_speculative_ngram_mod {
+    int32_t n_match = 24;
+
+    int32_t n_max = 64;
+    int32_t n_min = 48;
+
+    // shared instance of the ngram container for all speculative decoding contexts
+    std::shared_ptr<common_ngram_mod> obj;
+};
+
+struct common_params_speculative_ngram_map {
+    uint16_t size_n   = 12; // ngram size for lookup
+    uint16_t size_m   = 48; // mgram size for speculative tokens
+    uint16_t min_hits = 1;  // minimum hits at ngram/mgram lookup for mgram to be proposed
+};
+
+struct common_params_speculative_ngram_cache {
+    std::string lookup_cache_static;  // path of static ngram cache file for lookup decoding
+    std::string lookup_cache_dynamic; // path of dynamic ngram cache file for lookup decoding
+};
+
+struct common_params_speculative {
+    // TODO: become a vector in order to support "chains of speculators"
+    common_speculative_type type = COMMON_SPECULATIVE_TYPE_NONE;
+
+    common_params_speculative_draft draft;
+
+    common_params_speculative_ngram_mod ngram_mod;
+    common_params_speculative_ngram_map ngram_simple;
+    common_params_speculative_ngram_map ngram_map_k;
+    common_params_speculative_ngram_map ngram_map_k4v;
+
+    common_params_speculative_ngram_cache ngram_cache;
 
     bool has_dft() const {
-        return !mparams_dft.path.empty() || !mparams_dft.hf_repo.empty();
+        return !draft.mparams.path.empty() || !draft.mparams.hf_repo.empty();
     }
 };
 
 struct common_params_vocoder {
     struct common_params_model model;
 
-    std::string speaker_file = ""; // speaker file path                                      // NOLINT
+    std::string speaker_file; // speaker file path
 
-    bool use_guide_tokens = false; // enable guide tokens to improve TTS accuracy            // NOLINT
+    bool use_guide_tokens = false; // enable guide tokens to improve TTS accuracy
 };
 
 struct common_params_diffusion {
@@ -433,8 +453,8 @@ struct common_params {
 
     enum llama_split_mode split_mode = LLAMA_SPLIT_MODE_LAYER; // how to split the model across GPUs
 
-    struct cpu_params cpuparams;
-    struct cpu_params cpuparams_batch;
+    common_cpu_params cpuparams;
+    common_cpu_params cpuparams_batch;
 
     ggml_backend_sched_eval_callback cb_eval = nullptr;
     void * cb_eval_user_data                 = nullptr;
@@ -678,7 +698,7 @@ std::string common_params_get_system_info(const common_params & params);
 
 bool parse_cpu_range(const std::string & range, bool(&boolmask)[GGML_MAX_N_THREADS]);
 bool parse_cpu_mask(const std::string & mask, bool(&boolmask)[GGML_MAX_N_THREADS]);
-void postprocess_cpu_params(cpu_params & cpuparams, const cpu_params * role_model = nullptr);
+void postprocess_cpu_params(common_cpu_params & cpuparams, const common_cpu_params * role_model = nullptr);
 bool set_process_priority(enum ggml_sched_priority prio);
 
 //
@@ -846,7 +866,7 @@ common_init_result_ptr common_init_from_params(common_params & params);
 
 struct llama_model_params     common_model_params_to_llama  (      common_params & params);
 struct llama_context_params   common_context_params_to_llama(const common_params & params);
-struct ggml_threadpool_params ggml_threadpool_params_from_cpu_params(const cpu_params & params);
+struct ggml_threadpool_params ggml_threadpool_params_from_cpu_params(const common_cpu_params & params);
 
 // clear LoRA adapters from context, then apply new list of adapters
 void common_set_adapter_lora(struct llama_context * ctx, std::vector<common_adapter_lora_info> & lora);

common/debug.cpp

@@ -1,9 +1,38 @@
 #include "debug.h"
 
+#include "common.h"
 #include "log.h"
 
 #include <cmath>
+#include <regex>
 #include <string>
+#include <vector>
+
+struct common_debug_cb_user_data::impl {
+    std::vector<uint8_t>    data;
+    std::vector<std::regex> tensor_filters;
+    bool                    abort_on_nan{false};
+};
+
+common_debug_cb_user_data::common_debug_cb_user_data() : pimpl(std::make_unique<impl>()) {}
+common_debug_cb_user_data::~common_debug_cb_user_data() = default;
+
+common_debug_cb_user_data::common_debug_cb_user_data(common_params & params, const std::vector<std::string> & filter_patterns, bool abort_on_nan)
+    : pimpl(std::make_unique<impl>())
+{
+    for (const auto & pattern : filter_patterns) {
+        try {
+            std::string anchored_pattern = "^" + pattern;
+            pimpl->tensor_filters.emplace_back(anchored_pattern, std::regex::optimize);
+        } catch (const std::regex_error & e) {
+            throw std::runtime_error("Invalid regex pattern '" + pattern + "': " + e.what());
+        }
+    }
+    pimpl->abort_on_nan = abort_on_nan;
+
+    params.cb_eval           = common_debug_cb_eval;
+    params.cb_eval_user_data = this;
+}
 
 static std::string common_ggml_ne_string(const ggml_tensor * t) {
     std::string str;
@@ -47,8 +76,7 @@ static float common_ggml_get_float_value(const uint8_t * data,
 
 #define INDENT "    "
 
-template <bool abort>
-void common_debug_print_tensor(uint8_t * data, ggml_type type, const int64_t * ne, const size_t * nb, int64_t n) {
+static void common_debug_print_tensor(uint8_t * data, ggml_type type, const int64_t * ne, const size_t * nb, int64_t n, bool abort_on_nan) {
     GGML_ASSERT(n > 0);
     float sum = 0;
     for (int64_t i3 = 0; i3 < ne[3]; i3++) {
@@ -94,7 +122,7 @@ void common_debug_print_tensor(uint8_t * data, ggml_type type, const int64_t * n
         LOG(INDENT "sum = %f\n", sum);
     }
 
-    if constexpr (abort) {
+    if (abort_on_nan) {
         if (std::isnan(sum)) {
             LOG("encountered NaN - aborting\n");
             exit(0);
@@ -112,8 +140,9 @@ void common_debug_print_tensor(uint8_t * data, ggml_type type, const int64_t * n
  * @param user_data user data to pass at each call back
  * @return true to receive data or continue the graph, false otherwise
  */
-template <bool abort_on_nan> bool common_debug_cb_eval(struct ggml_tensor * t, bool ask, void * user_data) {
-    auto * cb_data = (base_callback_data *) user_data;
+bool common_debug_cb_eval(struct ggml_tensor * t, bool ask, void * user_data) {
+    auto * cb_data = (common_debug_cb_user_data *) user_data;
+    auto * pimpl = cb_data->pimpl.get();
 
     const struct ggml_tensor * src0 = t->src[0];
     const struct ggml_tensor * src1 = t->src[1];
@@ -122,10 +151,10 @@ template <bool abort_on_nan> bool common_debug_cb_eval(struct ggml_tensor * t, b
         return true;  // Always retrieve data
     }
 
-    bool matches_filter = cb_data->tensor_filters.empty();
+    bool matches_filter = pimpl->tensor_filters.empty();
 
     if (!matches_filter) {
-        for (const auto & filter : cb_data->tensor_filters) {
+        for (const auto & filter : pimpl->tensor_filters) {
             if (std::regex_search(t->name, filter)) {
                 matches_filter = true;
                 break;
@@ -148,20 +177,14 @@ template <bool abort_on_nan> bool common_debug_cb_eval(struct ggml_tensor * t, b
 
     if (!is_host) {
         auto n_bytes = ggml_nbytes(t);
-        cb_data->data.resize(n_bytes);
-        ggml_backend_tensor_get(t, cb_data->data.data(), 0, n_bytes);
+        pimpl->data.resize(n_bytes);
+        ggml_backend_tensor_get(t, pimpl->data.data(), 0, n_bytes);
     }
 
     if (!ggml_is_quantized(t->type) && matches_filter) {
-        uint8_t * data = is_host ? (uint8_t *) t->data : cb_data->data.data();
-        common_debug_print_tensor<abort_on_nan>(data, t->type, t->ne, t->nb, 3);
+        uint8_t * data = is_host ? (uint8_t *) t->data : pimpl->data.data();
+        common_debug_print_tensor(data, t->type, t->ne, t->nb, 3, pimpl->abort_on_nan);
     }
 
     return true;
 }
-
-// Explicit template instantiations
-template bool common_debug_cb_eval<false>(ggml_tensor *, bool, void *);
-template bool common_debug_cb_eval<true>(ggml_tensor *, bool, void *);
-template void common_debug_print_tensor<false>(uint8_t *, ggml_type, const int64_t *, const size_t *, int64_t);
-template void common_debug_print_tensor<true>(uint8_t *, ggml_type, const int64_t *, const size_t *, int64_t);

common/debug.h

@@ -1,43 +1,31 @@
 #pragma once
-#include "common.h"
+
+#include <memory>
 #include <string>
 #include <vector>
-#include <regex>
 
 // common debug functions and structs
 
-// Print a tensor's detailed data
-// data - the tensor's data in byte format
-// type - the tensor's quantization type
-// ne   - the tensor dimensions array
-// nb   - the tensor strides array
-// n    - the number of rows/columns to fully print
-template <bool abort_on_nan> void common_debug_print_tensor(uint8_t * data, ggml_type type, const int64_t * ne, const size_t * nb, int64_t n);
+struct common_params;
 
 // Intended to use as callback for ggml_backend_sched_eval_callback
 // prints tensors that are processed in the computation graph
-// by default prints all tensors, but can be configured by creating a `base_callback_data` instance with
-// non-empty filter_patterns. See examples/debug.ccp for possible usage patterns
-// The template parameter determines whether an error should be thrown whenever a NaN is encountered
+// by default prints all tensors, but can be configured by creating a `common_debug_cb_user_data` instance with
+// non-empty filter_patterns. See examples/debug.cpp for possible usage patterns
+// `common_debug_cb_user_data` contains `abort_on_nan` flag that determines whether an error should be thrown whenever a NaN is encountered
 // in a tensor (useful for stopping debug sessions on first erroneous tensor)
 // The callback data will be passed as the third parameter (user_data)
-template <bool abort_on_nan> bool common_debug_cb_eval(struct ggml_tensor * t, bool ask, void * user_data);
-struct base_callback_data {
-    std::vector<uint8_t>    data;
-    std::vector<std::regex> tensor_filters;
+bool common_debug_cb_eval(struct ggml_tensor * t, bool ask, void * user_data);
 
-    base_callback_data() = default;
+struct common_debug_cb_user_data {
+    struct impl;
+    std::unique_ptr<impl> pimpl;
 
-    base_callback_data(common_params & params, const std::vector<std::string> & filter_patterns) {
-        for (const auto & pattern : filter_patterns) {
-            try {
-                std::string anchored_pattern = "^" + pattern;
-                tensor_filters.emplace_back(anchored_pattern, std::regex::optimize);
-            } catch (const std::regex_error & e) {
-                throw std::runtime_error("Invalid regex pattern '" + pattern + "': " + e.what());
-            }
-        }
-        params.cb_eval           = common_debug_cb_eval<false>;
-        params.cb_eval_user_data = this;
-    }
+    common_debug_cb_user_data();
+    ~common_debug_cb_user_data();
+
+    common_debug_cb_user_data(const common_debug_cb_user_data &) = delete;
+    common_debug_cb_user_data & operator=(const common_debug_cb_user_data &) = delete;
+
+    common_debug_cb_user_data(common_params & params, const std::vector<std::string> & filter_patterns, bool abort_on_nan = false);
 };

common/download.cpp

@@ -627,7 +627,7 @@ static hf_cache::hf_file find_best_model(const hf_cache::hf_files & files,
     if (!tag.empty()) {
         tags.push_back(tag);
     } else {
-        tags = {"Q4_K_M", "Q4_0"};
+        tags = {"Q4_K_M", "Q8_0"};
     }
 
     for (const auto & t : tags) {

common/fit.cpp

@@ -856,7 +856,7 @@ void common_memory_breakdown_print(const struct llama_context * ctx) {
         ggml_backend_dev_memory(dev, &free, &total);
 
         const size_t self = mb.model + mb.context + mb.compute;
-        const size_t unaccounted = total - self - free;
+        const int64_t unaccounted = static_cast<int64_t>(total) - static_cast<int64_t>(free) - static_cast<int64_t>(self);
 
         table_data.push_back({
             template_gpu,
@@ -867,7 +867,7 @@ void common_memory_breakdown_print(const struct llama_context * ctx) {
             std::to_string(mb.model / MiB),
             std::to_string(mb.context / MiB),
             std::to_string(mb.compute / MiB),
-            std::to_string(unaccounted / MiB)});
+            std::to_string(unaccounted / static_cast<int64_t>(MiB))});
     }
 
     // print memory breakdown for host:

common/hf-cache.cpp

@@ -57,7 +57,7 @@ static fs::path get_cache_directory() {
 #ifndef _WIN32
         const struct passwd * pw = getpwuid(getuid());
 
-        if (pw->pw_dir && *pw->pw_dir) {
+        if (pw && pw->pw_dir && *pw->pw_dir) {
             return fs::path(pw->pw_dir) / ".cache" / "huggingface" / "hub";
         }
 #endif

common/jinja/runtime.h

@@ -106,10 +106,16 @@ struct statement {
     size_t pos; // position in source, for debugging
     virtual ~statement() = default;
     virtual std::string type() const { return "Statement"; }
+
     // execute_impl must be overridden by derived classes
-    virtual value execute_impl(context &) { throw std::runtime_error("cannot exec " + type()); }
+    virtual value execute_impl(context &) { throw_exec_error(); }
     // execute is the public method to execute a statement with error handling
     value execute(context &);
+
+private:
+    [[noreturn]] void throw_exec_error() const {
+        throw std::runtime_error("cannot exec " + type());
+    }
 };
 
 // Type Checking Utilities
@@ -143,7 +149,7 @@ struct program : public statement {
     program() = default;
     explicit program(statements && body) : body(std::move(body)) {}
     std::string type() const override { return "Program"; }
-    value execute_impl(context &) override {
+    [[noreturn]] value execute_impl(context &) override {
         throw std::runtime_error("Cannot execute program directly, use jinja::runtime instead");
     }
 };
@@ -195,7 +201,7 @@ struct break_statement : public statement {
         }
     };
 
-    value execute_impl(context &) override {
+    [[noreturn]] value execute_impl(context &) override {
         throw break_statement::signal();
     }
 };
@@ -209,7 +215,7 @@ struct continue_statement : public statement {
         }
     };
 
-    value execute_impl(context &) override {
+    [[noreturn]] value execute_impl(context &) override {
         throw continue_statement::signal();
     }
 };
@@ -509,7 +515,7 @@ struct slice_expression : public expression {
         chk_type<expression>(this->step_expr);
     }
     std::string type() const override { return "SliceExpression"; }
-    value execute_impl(context &) override {
+    [[noreturn]] value execute_impl(context &) override {
         throw std::runtime_error("must be handled by MemberExpression");
     }
 };

common/jinja/value.cpp

@@ -590,6 +590,10 @@ static bool string_endswith(const std::string & str, const std::string & suffix)
     return str.compare(str.length() - suffix.length(), suffix.length(), suffix) == 0;
 }
 
+[[noreturn]] static value string_join_not_implemented(const func_args &) {
+    throw not_implemented_exception("String join builtin not implemented");
+}
+
 const func_builtins & value_string_t::get_builtins() const {
     static const func_builtins builtins = {
         {"default", default_value},
@@ -851,9 +855,7 @@ const func_builtins & value_string_t::get_builtins() const {
             res->val_str.mark_input_based_on(val_input->as_string());
             return res;
         }},
-        {"join", [](const func_args &) -> value {
-            throw not_implemented_exception("String join builtin not implemented");
-        }},
+        {"join", string_join_not_implemented},
     };
     return builtins;
 }
@@ -884,6 +886,9 @@ const func_builtins & value_bool_t::get_builtins() const {
     return builtins;
 }
 
+[[noreturn]] static value array_unique_not_implemented(const func_args &) {
+    throw not_implemented_exception("Array unique builtin not implemented");
+}
 
 const func_builtins & value_array_t::get_builtins() const {
     static const func_builtins builtins = {
@@ -1084,13 +1089,14 @@ const func_builtins & value_array_t::get_builtins() const {
             std::reverse(arr.begin(), arr.end());
             return is_val<value_tuple>(val) ? mk_val<value_tuple>(std::move(arr)) : mk_val<value_array>(std::move(arr));
         }},
-        {"unique", [](const func_args &) -> value {
-            throw not_implemented_exception("Array unique builtin not implemented");
-        }},
+        {"unique", array_unique_not_implemented},
     };
     return builtins;
 }
 
+[[noreturn]] static value object_join_not_implemented(const func_args &) {
+    throw not_implemented_exception("object join not implemented");
+}
 
 const func_builtins & value_object_t::get_builtins() const {
     if (!has_builtins) {
@@ -1183,9 +1189,7 @@ const func_builtins & value_object_t::get_builtins() const {
             });
             return result;
         }},
-        {"join", [](const func_args &) -> value {
-            throw not_implemented_exception("object join not implemented");
-        }},
+        {"join", object_join_not_implemented},
     };
     return builtins;
 }

common/jinja/value.h

@@ -129,27 +129,25 @@ struct value_t {
     // Note: only for debugging and error reporting purposes
     virtual std::string type() const { return ""; }
 
-    virtual int64_t as_int() const { throw std::runtime_error(type() + " is not an int value"); }
-    virtual double as_float() const { throw std::runtime_error(type() + " is not a float value"); }
-    virtual string as_string() const { throw std::runtime_error(type() + " is not a string value"); }
-    virtual bool as_bool() const { throw std::runtime_error(type() + " is not a bool value"); }
-    virtual const std::vector<value> & as_array() const { throw std::runtime_error(type() + " is not an array value"); }
-    virtual const std::vector<std::pair<value, value>> & as_ordered_object() const { throw std::runtime_error(type() + " is not an object value"); }
-    virtual value invoke(const func_args &) const { throw std::runtime_error(type() + " is not a function value"); }
+    virtual int64_t as_int() const { throw_type_error("is not an int value"); }
+    virtual double as_float() const { throw_type_error("is not a float value"); }
+    virtual string as_string() const { throw_type_error("is not a string value"); }
+    virtual bool as_bool() const { throw_type_error("is not a bool value"); }
+    virtual const std::vector<value> & as_array() const { throw_type_error("is not an array value"); }
+    virtual const std::vector<std::pair<value, value>> & as_ordered_object() const { throw_type_error("is not an object value"); }
+    virtual value invoke(const func_args &) const { throw_type_error("is not a function value"); }
     virtual bool is_none() const { return false; }
     virtual bool is_undefined() const { return false; }
-    virtual const func_builtins & get_builtins() const {
-        throw std::runtime_error("No builtins available for type " + type());
-    }
+    virtual const func_builtins & get_builtins() const { throw_type_error("has no builtins"); }
 
-    virtual bool has_key(const value &) { throw std::runtime_error(type() + " is not an object value"); }
-    virtual void insert(const value & /* key */, const value & /* val */) { throw std::runtime_error(type() + " is not an object value"); }
-    virtual value & at(const value & /* key */, value & /* default_val */) { throw std::runtime_error(type() + " is not an object value"); }
-    virtual value & at(const value & /* key */) { throw std::runtime_error(type() + " is not an object value"); }
-    virtual value & at(const std::string & /* key */, value & /* default_val */) { throw std::runtime_error(type() + " is not an object value"); }
-    virtual value & at(const std::string & /* key */) { throw std::runtime_error(type() + " is not an object value"); }
-    virtual value & at(int64_t /* idx */, value & /* default_val */) { throw std::runtime_error(type() + " is not an array value"); }
-    virtual value & at(int64_t /* idx */) { throw std::runtime_error(type() + " is not an array value"); }
+    virtual bool has_key(const value &) { throw_type_error("is not an object value"); }
+    virtual void insert(const value & /* key */, const value & /* val */) { throw_type_error("is not an object value"); }
+    virtual value & at(const value & /* key */, value & /* default_val */) { throw_type_error("is not an object value"); }
+    virtual value & at(const value & /* key */) { throw_type_error("is not an object value"); }
+    virtual value & at(const std::string & /* key */, value & /* default_val */) { throw_type_error("is not an object value"); }
+    virtual value & at(const std::string & /* key */) { throw_type_error("is not an object value"); }
+    virtual value & at(int64_t /* idx */, value & /* default_val */) { throw_type_error("is not an array value"); }
+    virtual value & at(int64_t /* idx */) { throw_type_error("is not an array value"); }
 
     virtual bool is_numeric() const { return false; }
     virtual bool is_hashable() const { return false; }
@@ -163,6 +161,11 @@ struct value_t {
     // Note: only for debugging purposes
     virtual std::string as_repr() const { return as_string().str(); }
 
+private:
+    [[noreturn]] void throw_type_error(const char* expected) const {
+        throw std::runtime_error(type() + " " + expected);
+    }
+
 protected:
     virtual bool equivalent(const value_t &) const = 0;
     virtual bool nonequal(const value_t & other) const { return !equivalent(other); }

common/log.cpp

@@ -49,7 +49,7 @@ enum common_log_col : int {
 };
 
 // disable colors by default
-static std::vector<const char *> g_col = {
+static const char* g_col[] = {
     "",
     "",
     "",
@@ -247,7 +247,6 @@ public:
 
             entries = std::move(new_entries);
         }
-
         cv.notify_one();
     }
 
@@ -265,7 +264,6 @@ public:
                 {
                     std::unique_lock<std::mutex> lock(mtx);
                     cv.wait(lock, [this]() { return head != tail; });
-
                     cur = entries[head];
 
                     head = (head + 1) % entries.size();
@@ -301,7 +299,6 @@ public:
 
                 tail = (tail + 1) % entries.size();
             }
-
             cv.notify_one();
         }
 
@@ -338,7 +335,7 @@ public:
             g_col[COMMON_LOG_COL_CYAN]    = LOG_COL_CYAN;
             g_col[COMMON_LOG_COL_WHITE]   = LOG_COL_WHITE;
         } else {
-            for (size_t i = 0; i < g_col.size(); i++) {
+            for (size_t i = 0; i < std::size(g_col); i++) {
                 g_col[i] = "";
             }
         }
@@ -368,14 +365,20 @@ struct common_log * common_log_init() {
 }
 
 struct common_log * common_log_main() {
-    static struct common_log log;
+    // We intentionally leak (i.e. do not delete) the logger singleton because
+    // common_log destructor called at DLL teardown phase will cause hanging on Windows.
+    // OS will release resources anyway so it should not be a significant issue,
+    // though this design may cause logs to be lost if not flushed before the program exits.
+    // Refer to https://github.com/ggml-org/llama.cpp/issues/22142 for details.
+    static struct common_log * log;
     static std::once_flag    init_flag;
     std::call_once(init_flag, [&]() {
+        log = new common_log;
         // Set default to auto-detect colors
-        log.set_colors(tty_can_use_colors());
+        log->set_colors(tty_can_use_colors());
     });
 
-    return &log;
+    return log;
 }
 
 void common_log_pause(struct common_log * log) {

common/log.h

@@ -49,7 +49,11 @@ void common_log_default_callback(enum ggml_log_level level, const char * text, v
 struct common_log;
 
 struct common_log * common_log_init();
-struct common_log * common_log_main(); // singleton, automatically destroys itself on exit
+
+// Singleton, intentionally leaked to avoid Windows teardown hangs.
+// Call common_log_flush() before exit if you want to ensure all logs are flushed.
+struct common_log * common_log_main();
+
 void                common_log_pause (struct common_log * log); // pause  the worker thread, not thread-safe
 void                common_log_resume(struct common_log * log); // resume the worker thread, not thread-safe
 void                common_log_free  (struct common_log * log);

common/preset.cpp

@@ -43,7 +43,7 @@ static std::set<std::string> get_remote_preset_whitelist(const std::map<std::str
     for (const auto & it : key_to_opt) {
         const std::string & key = it.first;
         const common_arg & opt = it.second;
-        if (allowed_options.find(key) != allowed_options.end() || opt.is_sparam) {
+        if (allowed_options.find(key) != allowed_options.end() || opt.is_sampling) {
             allowed_keys.insert(key);
             // also add variant keys (args without leading dashes and env vars)
             for (const auto & arg : opt.get_args()) {

common/reasoning-budget.cpp

@@ -122,6 +122,20 @@ static void common_reasoning_budget_accept(struct llama_sampler * smpl, llama_to
             }
             break;
         case REASONING_BUDGET_DONE:
+            // Re-arm on a new start tag: some models emit multiple <think> blocks
+            // per response, and each should get a fresh budget window.
+            if (ctx->start_matcher.advance(token)) {
+                ctx->state = REASONING_BUDGET_COUNTING;
+                ctx->remaining = ctx->budget;
+                ctx->end_matcher.reset();
+                LOG_INF("reasoning-budget: re-activated on new start tag, budget=%d tokens\n", ctx->budget);
+
+                if (ctx->remaining <= 0) {
+                    ctx->state = REASONING_BUDGET_FORCING;
+                    ctx->force_pos = 0;
+                    LOG_INF("reasoning-budget: budget=0, forcing immediately\n");
+                }
+            }
             break;
     }
 }
@@ -218,34 +232,6 @@ static struct llama_sampler * common_reasoning_budget_init_state(
     );
 }
 
-struct llama_sampler * common_reasoning_budget_init(
-        const struct llama_vocab       * vocab,
-        const std::vector<llama_token> & start_tokens,
-        const std::vector<llama_token> & end_tokens,
-        const std::vector<llama_token> & forced_tokens,
-        int32_t                          budget,
-        const std::vector<llama_token> & prefill_tokens) {
-    // Determine initial state from prefill: COUNTING if the prefill begins with
-    // the start sequence but does not also contain the end sequence after it.
-    common_reasoning_budget_state initial_state = REASONING_BUDGET_IDLE;
-    if (!prefill_tokens.empty() && !start_tokens.empty() &&
-            prefill_tokens.size() >= start_tokens.size() &&
-            std::equal(start_tokens.begin(), start_tokens.end(), prefill_tokens.begin())) {
-        initial_state = REASONING_BUDGET_COUNTING;
-        // If the end sequence also follows the start in the prefill, reasoning
-        // was opened and immediately closed — stay IDLE.
-        if (!end_tokens.empty() &&
-                prefill_tokens.size() >= start_tokens.size() + end_tokens.size()) {
-            auto end_start = prefill_tokens.end() - (ptrdiff_t) end_tokens.size();
-            if (end_start >= prefill_tokens.begin() + (ptrdiff_t) start_tokens.size() &&
-                    std::equal(end_tokens.begin(), end_tokens.end(), end_start)) {
-                initial_state = REASONING_BUDGET_IDLE;
-            }
-        }
-    }
-    return common_reasoning_budget_init_state(vocab, start_tokens, end_tokens, forced_tokens, budget, initial_state);
-}
-
 struct llama_sampler * common_reasoning_budget_init(
         const struct llama_vocab       * vocab,
         const std::vector<llama_token> & start_tokens,

common/reasoning-budget.h

@@ -29,10 +29,7 @@ enum common_reasoning_budget_state {
 //   end_tokens     - token sequence for natural deactivation
 //   forced_tokens  - token sequence forced when budget expires
 //   budget         - max tokens allowed in the reasoning block
-//   prefill_tokens - tokens already present in the prompt (generation prompt);
-//                    used to determine the initial state: COUNTING if they begin
-//                    with start_tokens (but don't also end with end_tokens),
-//                    IDLE otherwise. COUNTING with budget <= 0 is promoted to FORCING.
+//   initial_state  - initial state
 //
 struct llama_sampler * common_reasoning_budget_init(
         const struct llama_vocab       * vocab,
@@ -40,16 +37,6 @@ struct llama_sampler * common_reasoning_budget_init(
         const std::vector<llama_token> & end_tokens,
         const std::vector<llama_token> & forced_tokens,
         int32_t                          budget,
-        const std::vector<llama_token> & prefill_tokens = {});
-
-// Variant that takes an explicit initial state (used by tests and clone).
-// COUNTING with budget <= 0 is promoted to FORCING.
-struct llama_sampler * common_reasoning_budget_init(
-        const struct llama_vocab       * vocab,
-        const std::vector<llama_token> & start_tokens,
-        const std::vector<llama_token> & end_tokens,
-        const std::vector<llama_token> & forced_tokens,
-        int32_t                          budget,
-        common_reasoning_budget_state    initial_state);
+        common_reasoning_budget_state    initial_state = REASONING_BUDGET_IDLE);
 
 common_reasoning_budget_state common_reasoning_budget_get_state(const struct llama_sampler * smpl);

common/sampling.cpp

@@ -260,32 +260,35 @@ struct common_sampler * common_sampler_init(const struct llama_model * model, st
         }
     }
 
+    // Compute prefill tokens from the generation prompt
+    std::vector<llama_token> prefill_tokens;
+    if (!params.generation_prompt.empty()) {
+        GGML_ASSERT(vocab != nullptr);
+        auto tokens = common_tokenize(vocab, params.generation_prompt, false, true);
+        for (size_t i = 0; i < tokens.size(); i++) {
+            std::string piece = common_token_to_piece(vocab, tokens[i], true);
+            if (i == 0 && std::isspace(piece[0]) && !std::isspace(params.generation_prompt[0])) {
+                // Some tokenizers will add a space before the first special token, need to exclude
+                continue;
+            }
+            LOG_DBG("%s: prefill token: %d = %s\n", __func__, tokens[i], piece.c_str());
+            prefill_tokens.push_back(tokens[i]);
+        }
+    }
+
     // Feed generation prompt tokens to the grammar sampler so it advances past
     // tokens the template already placed in the prompt.
     // Only applies to output-format and tool-call grammars; user-supplied grammars must not be prefilled.
-    std::vector<llama_token> prefill_tokens;
-    if (!params.generation_prompt.empty() && common_grammar_needs_prefill(params.grammar)) {
-        GGML_ASSERT(vocab != nullptr);
-        prefill_tokens = common_tokenize(vocab, params.generation_prompt, false, true);
-        if (!prefill_tokens.empty()) {
-            std::string first_token = common_token_to_piece(vocab, prefill_tokens[0], true);
-            if (std::isspace(first_token[0]) && !std::isspace(params.generation_prompt[0])) {
-                // Some tokenizers will add a space before the first special token, need to remove
-                prefill_tokens = std::vector<llama_token>(prefill_tokens.begin() + 1, prefill_tokens.end());
-            }
-        }
-
-        if (grmr && !params.grammar_lazy) {
-            try {
-                for (const auto & token : prefill_tokens) {
-                    llama_sampler_accept(grmr, token);
-                    LOG_DBG("%s: accepted prefill token (%d)\n", __func__, token);
-                }
-            } catch (std::exception &e) {
-                LOG_ERR("%s: error initializing grammar sampler for grammar:\n%s\n\nGeneration prompt:\n'%s'\n", __func__,
-                    common_grammar_value(params.grammar).c_str(), params.generation_prompt.c_str());
-                throw e;
+    if (grmr && !params.grammar_lazy && common_grammar_needs_prefill(params.grammar)) {
+        try {
+            for (const auto & token : prefill_tokens) {
+                llama_sampler_accept(grmr, token);
+                LOG_DBG("%s: grammar accepted prefill token (%d)\n", __func__, token);
             }
+        } catch (std::exception &e) {
+            LOG_ERR("%s: error initializing grammar sampler for grammar:\n%s\n\nGeneration prompt:\n'%s'\n", __func__,
+                common_grammar_value(params.grammar).c_str(), params.generation_prompt.c_str());
+            throw e;
         }
     }
 
@@ -296,8 +299,12 @@ struct common_sampler * common_sampler_init(const struct llama_model * model, st
             params.reasoning_budget_start,
             params.reasoning_budget_end,
             params.reasoning_budget_forced,
-            params.reasoning_budget_tokens < 0 ? INT_MAX : params.reasoning_budget_tokens,
-            prefill_tokens);
+            params.reasoning_budget_tokens < 0 ? INT_MAX : params.reasoning_budget_tokens);
+
+        for (const auto & token : prefill_tokens) {
+            llama_sampler_accept(rbudget, token);
+            LOG_DBG("%s: reasoning-budget accepted prefill token (%d)\n", __func__, token);
+        }
     }
 
     if (params.has_logit_bias()) {
@@ -431,7 +438,7 @@ static bool grammar_should_apply(struct common_sampler * gsmpl) {
     return true;
 }
 
-void common_sampler_accept(struct common_sampler * gsmpl, llama_token token, bool accept_grammar) {
+void common_sampler_accept(struct common_sampler * gsmpl, llama_token token, bool is_generated) {
     if (!gsmpl) {
         return;
     }
@@ -439,9 +446,11 @@ void common_sampler_accept(struct common_sampler * gsmpl, llama_token token, boo
     const auto tm = gsmpl->tm();
 
     // grammar_should_apply() checks the reasoning budget state, so calculate this before we accept
-    accept_grammar = accept_grammar && grammar_should_apply(gsmpl);
+    const auto accept_grammar = is_generated && grammar_should_apply(gsmpl);
 
-    llama_sampler_accept(gsmpl->rbudget, token);
+    if (gsmpl->rbudget && is_generated) {
+        llama_sampler_accept(gsmpl->rbudget, token);
+    }
 
     if (gsmpl->grmr && accept_grammar) {
         llama_sampler_accept(gsmpl->grmr, token);

common/sampling.h

@@ -41,8 +41,8 @@ struct common_sampler * common_sampler_init(const struct llama_model * model, st
 
 void common_sampler_free(struct common_sampler * gsmpl);
 
-// if accept_grammar is true, the token is accepted both by the sampling chain and the grammar
-void                    common_sampler_accept(struct common_sampler * gsmpl, llama_token token, bool accept_grammar);
+// if is_generated is true, the token is accepted by the sampling chain, the reasoning budget sampler, and the grammar sampler
+void                    common_sampler_accept(struct common_sampler * gsmpl, llama_token token, bool is_generated);
 void                    common_sampler_reset (struct common_sampler * gsmpl);
 struct common_sampler * common_sampler_clone (struct common_sampler * gsmpl);
 

common/speculative.cpp

@@ -61,18 +61,26 @@ static bool common_speculative_are_compatible(
     LOG_DBG("%s: vocab_type dft: %d\n", __func__, vocab_type_dft);
 
     if (vocab_type_tgt != vocab_type_dft) {
-        LOG_DBG("%s: draft model vocab type must match target model to use speculation but ", __func__);
-        LOG_DBG("vocab_type_dft = %d while vocab_type_tgt = %d\n", vocab_type_dft, vocab_type_tgt);
+        LOG_WRN("%s: draft model vocab type must match target model to use speculation but "
+                "vocab_type_dft = %d while vocab_type_tgt = %d\n", __func__, vocab_type_dft, vocab_type_tgt);
         return false;
     }
 
-    if (
-        llama_vocab_get_add_bos(vocab_tgt) != llama_vocab_get_add_bos(vocab_dft) ||
-        llama_vocab_get_add_eos(vocab_tgt) != llama_vocab_get_add_eos(vocab_dft) ||
-        llama_vocab_bos(vocab_tgt) != llama_vocab_bos(vocab_dft) ||
-        llama_vocab_eos(vocab_tgt) != llama_vocab_eos(vocab_dft)
-    ) {
-        LOG_DBG("%s: draft model special tokens must match target model to use speculation\n", __func__);
+    if (llama_vocab_get_add_bos(vocab_tgt) != llama_vocab_get_add_bos(vocab_dft) ||
+        (llama_vocab_get_add_bos(vocab_tgt) && llama_vocab_bos(vocab_tgt) != llama_vocab_bos(vocab_dft))) {
+        LOG_WRN("%s: draft model bos tokens must match target model to use speculation. add: %d - %d, id: %d - %d)\n",
+                __func__,
+                llama_vocab_get_add_bos(vocab_tgt), llama_vocab_get_add_bos(vocab_dft),
+                llama_vocab_bos(vocab_tgt), llama_vocab_bos(vocab_dft));
+        return false;
+    }
+
+    if (llama_vocab_get_add_eos(vocab_tgt) != llama_vocab_get_add_eos(vocab_dft) ||
+        (llama_vocab_get_add_eos(vocab_tgt) && llama_vocab_eos(vocab_tgt) != llama_vocab_eos(vocab_dft))) {
+        LOG_WRN("%s: draft model eos tokens must match target model to use speculation. add: %d - %d, id: %d - %d)\n",
+                __func__,
+                llama_vocab_get_add_eos(vocab_tgt), llama_vocab_get_add_eos(vocab_dft),
+                llama_vocab_eos(vocab_tgt), llama_vocab_eos(vocab_dft));
         return false;
     }
 
@@ -143,6 +151,9 @@ struct common_speculative_state {
             llama_tokens & result) = 0;
 
     virtual void accept(uint16_t n_accepted) = 0;
+
+    virtual int32_t n_max(const common_params_speculative & params) const = 0;
+    virtual int32_t n_min(const common_params_speculative & params) const = 0;
 };
 
 struct common_speculative_checkpoint {
@@ -156,8 +167,6 @@ struct common_speculative_checkpoint {
     size_t size() const {
         return data.size();
     }
-
-    size_t ckpt_size   = 0;
 };
 
 struct common_speculative_state_draft : public common_speculative_state {
@@ -165,7 +174,7 @@ struct common_speculative_state_draft : public common_speculative_state {
     llama_context * ctx_dft;
 
     bool use_ckpt = false;
-    struct common_speculative_checkpoint ckpt;
+    common_speculative_checkpoint ckpt;
 
     common_sampler * smpl;
 
@@ -238,26 +247,16 @@ struct common_speculative_state_draft : public common_speculative_state {
         llama_batch_free(batch);
     }
 
-    void begin(const llama_tokens & prompt) override {
-        if (use_ckpt && ckpt.size() > 0) {
-            // delete checkpoint
-            LOG_DBG("%s: delete checkpoint, prompt.size=%zu, pos_min=%d, pos_max=%d, n_tokens=%" PRId64 ", size=%.3f MiB\n",
-                    __func__, prompt.size(), ckpt.pos_min, ckpt.pos_max, ckpt.n_tokens, (float) ckpt.data.size() / 1024 / 1024);
-            ckpt.pos_min   = 0;
-            ckpt.pos_max   = 0;
-            ckpt.n_tokens  = 0;
-            ckpt.ckpt_size = 0;
-            ckpt.data.clear();
-        }
+    void begin(const llama_tokens & /*prompt*/) override {
     }
 
-    size_t draft_create_checkpoint(int n_tokens_prompt, int n_tokens_batch) {
+    size_t create_checkpoint(int n_tokens_prompt) {
         int slot_id = 0;
         const size_t checkpoint_size = llama_state_seq_get_size_ext(ctx_dft, slot_id, LLAMA_STATE_SEQ_FLAGS_PARTIAL_ONLY);
 
         ckpt.pos_min  = llama_memory_seq_pos_min(llama_get_memory(ctx_dft), slot_id);
         ckpt.pos_max  = llama_memory_seq_pos_max(llama_get_memory(ctx_dft), slot_id);
-        ckpt.n_tokens = n_tokens_prompt - n_tokens_batch;
+        ckpt.n_tokens = n_tokens_prompt;
         ckpt.data.resize(checkpoint_size);
 
         const size_t n = llama_state_seq_get_data_ext(ctx_dft, ckpt.data.data(), checkpoint_size, slot_id, LLAMA_STATE_SEQ_FLAGS_PARTIAL_ONLY);
@@ -270,13 +269,13 @@ struct common_speculative_state_draft : public common_speculative_state {
         return n;
     }
 
-    size_t draft_restore_checkpoint(size_t ckpt_size_part_expected) {
+    size_t restore_checkpoint() {
         int slot_id = 0;
         LOG_DBG("%s: pos_min = %d, pos_max = %d\n", __func__, ckpt.pos_min, ckpt.pos_max);
         const size_t n = llama_state_seq_set_data_ext(ctx_dft, ckpt.data.data(), ckpt.size(), slot_id, LLAMA_STATE_SEQ_FLAGS_PARTIAL_ONLY);
-        if (n != ckpt_size_part_expected) {
-            GGML_ABORT("%s: failed to restore context checkpoint (pos_min=%d, pos_max=%d, size=%zu, get_data_ext->%zu, set_data_ext->%zu",
-                        __func__, ckpt.pos_min, ckpt.pos_max, ckpt.size(), ckpt_size_part_expected, n);
+        if (n != ckpt.size()) {
+            GGML_ABORT("%s: failed to restore context checkpoint (pos_min=%d, pos_max=%d, size=%zu",
+                        __func__, ckpt.pos_min, ckpt.pos_max, ckpt.size());
         }
         llama_memory_seq_rm(llama_get_memory(ctx_dft), slot_id, ckpt.pos_max + 1, -1);
 
@@ -288,6 +287,8 @@ struct common_speculative_state_draft : public common_speculative_state {
             const llama_tokens & prompt_tgt,
             llama_token id_last,
             llama_tokens & result) override {
+        const auto & sparams = params.draft;
+
         auto * spec = this;
 
         auto & batch      = spec->batch;
@@ -301,7 +302,7 @@ struct common_speculative_state_draft : public common_speculative_state {
         int reuse_i = 0; // index of part to be reused in prompt_dft
         int reuse_n = 0; // length of part to be reused in prompt_dft
 
-        const int n_ctx = llama_n_ctx(ctx_dft) - params.n_max;
+        const int n_ctx = llama_n_ctx(ctx_dft) - sparams.n_max;
 
         llama_tokens prompt_cnv;
         if (!spec->vocab_cmpt) {
@@ -333,13 +334,18 @@ struct common_speculative_state_draft : public common_speculative_state {
 
         const int i_start = std::max<int>(0, (int) prompt_cur.size() - n_ctx);
 
+        if (use_ckpt && i_start > 0) {
+            LOG_WRN("%s: context shift is not supported with checkpoint-based contexts - skipping\n", __func__);
+            return;
+        }
+
         // reuse as much as possible from the old draft context
         // ideally, the draft context should be as big as the target context and we will always reuse the entire prompt
         for (int i = 0; i < (int) prompt_dft.size(); ++i) {
             int cur = 0;
             while (i_start + cur < (int) prompt_cur.size() &&
-                    i       + cur < (int) prompt_dft.size() &&
-                    prompt_cur[i_start + cur] == prompt_dft[i + cur]) {
+                   i       + cur < (int) prompt_dft.size() &&
+                   prompt_cur[i_start + cur] == prompt_dft[i + cur]) {
                 cur++;
             }
 
@@ -347,21 +353,26 @@ struct common_speculative_state_draft : public common_speculative_state {
                 reuse_i = i;
                 reuse_n = cur;
             }
+
+            if (use_ckpt) {
+                break;
+            }
         }
 
         LOG_DBG("%s: reuse_i = %d, reuse_n = %d, #prompt_dft = %zu, #prompt_cur = %zu\n",
                 __func__, reuse_i, reuse_n, prompt_dft.size(), prompt_cur.size());
-        if (use_ckpt && ckpt.ckpt_size == 0 && reuse_n > 0) {
-            LOG_DBG("%s: no checkpoint available, no reuse, (reuse_i=%d, reuse_n=%d) -> (0, 0)\n",
-                    __func__, reuse_i, reuse_n);
+        if (use_ckpt && ckpt.n_tokens > reuse_n) {
+            LOG_DBG("%s: checkpoint (n_tokens = %d) is outdated -> delete it\n", __func__, (int) ckpt.n_tokens);
+
             reuse_i = 0;
             reuse_n = 0;
+
+            ckpt = {};
         }
 
         result.clear();
-        result.reserve(params.n_max);
+        result.reserve(sparams.n_max);
 
-        bool needs_ckpt = use_ckpt && prompt_dft.size() > 0;
         if (reuse_n == 0 || (use_ckpt && reuse_i > 0)) {
             llama_memory_clear(mem_dft, false);
             prompt_dft.clear();
@@ -372,7 +383,7 @@ struct common_speculative_state_draft : public common_speculative_state {
                 for (int i = reuse_i + reuse_n + 1; i < (int) prompt_dft.size(); ++i) {
                     result.push_back(prompt_dft[i]);
 
-                    if (params.n_max <= (int) result.size()) {
+                    if (sparams.n_max <= (int) result.size()) {
                         break;
                     }
                 }
@@ -380,50 +391,38 @@ struct common_speculative_state_draft : public common_speculative_state {
                 return;
             }
 
-            bool do_restore = false;
-            if (prompt_dft.size() > prompt_cur.size() && reuse_i + reuse_n < (int64_t) prompt_dft.size()) {
-                // This can happen after a partial acceptance (speculative decoding with checkpoints)
-                LOG_DBG("%s: #prompt_dft=%zu, #prompt_cur=%zu, shorten draft\n",
-                        __func__, prompt_dft.size(), prompt_cur.size());
-                prompt_dft.resize(prompt_cur.size());
-                do_restore = true;
-            }
-
             if (reuse_i > 0) {
+                GGML_ASSERT(!use_ckpt);
+
                 bool is_removed = llama_memory_seq_rm (mem_dft, 0, 0, reuse_i);
                 if (!is_removed) {
                     LOG_ERR("%s: llama_memory_seq_rm failed, reuse_i=%d\n", __func__, reuse_i);
+                    return;
                 }
                 llama_memory_seq_add(mem_dft, 0, reuse_i, -1, -reuse_i);
 
                 prompt_dft.erase(prompt_dft.begin(), prompt_dft.begin() + reuse_i);
             }
 
-            if (reuse_n < (int) prompt_dft.size() || do_restore) {
+            if (reuse_n < (int) prompt_dft.size()) {
                 if (use_ckpt) {
-                    if (ckpt.n_tokens > (int64_t) prompt_dft.size()) {
-                        LOG_INF("%s: checkpoint is too large, prompt_tgt.size=%zu, ckpt.n_tokens=%" PRId64 ", reuse_n=%d, prompt_dft.size=%zu\n",
-                                __func__, prompt_tgt.size(), ckpt.n_tokens, reuse_n, prompt_dft.size());
+                    if (ckpt.n_tokens > 0) {
+                        LOG_DBG("%s: restoring checkpoint, reuse_n=%d, prompt_dft.size=%zu\n", __func__, reuse_n, prompt_dft.size());
+                        restore_checkpoint();
+                        reuse_n = ckpt.n_tokens;
+                        prompt_dft.resize(reuse_n);
                     }
-                    draft_restore_checkpoint(ckpt.ckpt_size);
-                    reuse_n = ckpt.n_tokens;
-                    prompt_dft.resize(reuse_n);
-                    needs_ckpt = false;
                 } else {
-                    bool is_removed = llama_memory_seq_rm (mem_dft, 0, reuse_n, -1);
+                    const bool is_removed = llama_memory_seq_rm(mem_dft, 0, reuse_n, -1);
                     if (!is_removed) {
-                        LOG_ERR("%s: llama_memory_seq_rm failed, reuse_n=%d, prompt_dft.size=%zu\n",
-                                __func__, reuse_n, prompt_dft.size());
+                        LOG_ERR("%s: llama_memory_seq_rm failed, reuse_n=%d, prompt_dft.size=%zu\n", __func__, reuse_n, prompt_dft.size());
+                        return;
                     }
                     prompt_dft.erase(prompt_dft.begin() + reuse_n, prompt_dft.end());
                 }
             }
         }
 
-        if (needs_ckpt) {
-            ckpt.ckpt_size = draft_create_checkpoint(prompt_dft.size(), batch.n_tokens);
-        }
-
         // prepare a batch to evaluate any new tokens in the prompt
         common_batch_clear(batch);
 
@@ -437,12 +436,17 @@ struct common_speculative_state_draft : public common_speculative_state {
         // we should rarely end-up here during normal decoding
         if (batch.n_tokens > 0) {
             //LOG_DBG("%s: draft prompt batch: %s\n", __func__, string_from(ctx, batch).c_str());
+            LOG_DBG("%s: draft prompt batch: %d tokens\n", __func__, batch.n_tokens);
 
             int ret = llama_decode(ctx_dft, batch);
             if (ret != 0 && ret != 1) {
                 LOG_WRN("%s: llama_decode returned %d, prompt_cur.size=%zu\n",
                         __func__, ret, prompt_cur.size());
             }
+
+            if (use_ckpt) {
+                create_checkpoint(prompt_dft.size());
+            }
         }
 
         const llama_pos n_past = prompt_dft.size();
@@ -454,7 +458,7 @@ struct common_speculative_state_draft : public common_speculative_state {
 
         prompt_dft.push_back(id_last);
 
-        LOG_DBG("%s: draft prompt: %s\n", __func__, string_from(ctx_dft, prompt_dft).c_str());
+        //LOG_DBG("%s: draft prompt: %s\n", __func__, string_from(ctx_dft, prompt_dft).c_str());
 
         int ret = llama_decode(ctx_dft, batch);
         if (ret != 0 && ret != 1) {
@@ -465,7 +469,7 @@ struct common_speculative_state_draft : public common_speculative_state {
         common_sampler_reset(smpl);
 
         // sample n_draft tokens from the draft model
-        for (int i = 0; i < params.n_max; ++i) {
+        for (int i = 0; i < sparams.n_max; ++i) {
             common_batch_clear(batch);
 
             common_sampler_sample(smpl, ctx_dft, 0, true);
@@ -482,14 +486,14 @@ struct common_speculative_state_draft : public common_speculative_state {
 
             common_sampler_accept(smpl, id, true);
 
-            result.push_back(id);
-
-            if (params.n_max <= (int) result.size()) {
+            // only collect very high-confidence draft tokens
+            if (cur_p->data[0].p < sparams.p_min) {
                 break;
             }
 
-            // only collect very high-confidence draft tokens
-            if (cur_p->data[0].p < params.p_min) {
+            result.push_back(id);
+
+            if (sparams.n_max <= (int) result.size()) {
                 break;
             }
 
@@ -510,10 +514,14 @@ struct common_speculative_state_draft : public common_speculative_state {
             detokenized = replace_to_tgt(detokenized);
             LOG_DBG("draft->main detokenized string: '%s'\n", detokenized.c_str());
             result = common_tokenize(ctx_tgt, detokenized, false, true);
-            if (result.size() > (size_t)params.n_max) {
-                result.resize(params.n_max);
+            if (result.size() > (size_t) sparams.n_max) {
+                result.resize(sparams.n_max);
             }
         }
+
+        if (result.size() < (size_t) sparams.n_min) {
+            result.clear();
+        }
     }
 
     void accept(uint16_t n_accepted) override {
@@ -521,6 +529,14 @@ struct common_speculative_state_draft : public common_speculative_state {
         GGML_UNUSED(n_accepted);
     }
 
+    int32_t n_max(const common_params_speculative & params) const override {
+        return params.draft.n_max;
+    }
+
+    int32_t n_min(const common_params_speculative & params) const override {
+        return params.draft.n_min;
+    }
+
     std::string replace_to_dft(const std::string & input) const {
         std::string result = input;
 
@@ -573,6 +589,14 @@ struct common_speculative_state_eagle3 : public common_speculative_state {
         // noop
         GGML_UNUSED(n_accepted);
     }
+
+    int32_t n_max(const common_params_speculative & params) const override {
+        return params.draft.n_max;
+    }
+
+    int32_t n_min(const common_params_speculative & params) const override {
+        return params.draft.n_min;
+    }
 };
 
 // state of self-speculation (simple implementation, not ngram-map)
@@ -602,19 +626,27 @@ struct common_speculative_state_ngram_simple : public common_speculative_state {
         // noop
         GGML_UNUSED(n_accepted);
     }
+
+    int32_t n_max(const common_params_speculative & /*params*/) const override {
+        return config.size_mgram;
+    }
+
+    int32_t n_min(const common_params_speculative & /*params*/) const override {
+        return config.size_mgram;
+    }
 };
 
 struct common_speculative_state_ngram_map_k : public common_speculative_state {
     // draft ngram map for speculative decoding without draft model
-    common_ngram_map map;
+    common_ngram_map config;
 
     common_speculative_state_ngram_map_k(
             enum common_speculative_type type,
-            common_ngram_map map)
-        : common_speculative_state(type), map(std::move(map)) {}
+            common_ngram_map config)
+        : common_speculative_state(type), config(std::move(config)) {}
 
     void begin(const llama_tokens & prompt) override {
-        common_ngram_map_begin(map, prompt);
+        common_ngram_map_begin(config, prompt);
     }
 
     void draft(
@@ -622,12 +654,20 @@ struct common_speculative_state_ngram_map_k : public common_speculative_state {
             const llama_tokens & prompt_tgt,
             llama_token id_last,
             llama_tokens & result) override {
-        common_ngram_map_draft(map, prompt_tgt, id_last, result);
+        common_ngram_map_draft(config, prompt_tgt, id_last, result);
         GGML_UNUSED(params);
     }
 
     void accept(uint16_t n_accepted) override {
-        common_ngram_map_accept(map, n_accepted);
+        common_ngram_map_accept(config, n_accepted);
+    }
+
+    int32_t n_max(const common_params_speculative & /*params*/) const override {
+        return config.size_value;
+    }
+
+    int32_t n_min(const common_params_speculative & /*params*/) const override {
+        return config.size_value;
     }
 };
 
@@ -684,7 +724,7 @@ struct common_speculative_state_ngram_mod : public common_speculative_state {
             const llama_tokens & prompt_tgt,
             llama_token id_last,
             llama_tokens & result) override {
-        GGML_UNUSED(params);
+        const auto & sparams = params.ngram_mod;
 
         n_draft_last = 0;
 
@@ -704,16 +744,16 @@ struct common_speculative_state_ngram_mod : public common_speculative_state {
             i_last = cur_len - n;
         }
 
-        result.resize(n + params.n_max);
+        result.resize(n + sparams.n_max);
         for (size_t i = 0; i < n - 1; ++i) {
             result[i] = prompt_tgt[cur_len - n + 1 + i];
         }
         result[n - 1] = id_last;
 
-        for (int i = 0; i < params.n_max; ++i) {
+        for (int i = 0; i < sparams.n_max; ++i) {
             const llama_token token = mod.get(result.data() + i);
             if (token == common_ngram_mod::EMPTY) {
-                if (i < params.n_min) {
+                if (i < sparams.n_min) {
                     result.clear();
                     return;
                 }
@@ -735,17 +775,15 @@ struct common_speculative_state_ngram_mod : public common_speculative_state {
     }
 
     void accept(uint16_t n_accepted) override {
-        if (verbose) {
-            LOG_INF("%s: accepted %d tokens from %zu drafted tokens\n", __func__, n_accepted, n_draft_last);
-        }
-
         // compute acceptance fraction if we have a recorded draft length
         if (n_draft_last > 0) {
             const double f_acc = (double)n_accepted / (double)n_draft_last;
             if (f_acc < 0.5) {
                 n_low++;
                 if (n_low >= 3) {
-                    LOG_WRN("%s: low acceptance streak (%d) – resetting ngram_mod\n", __func__, n_low);
+                    if (verbose) {
+                        LOG_WRN("%s: low acceptance streak (%d) – resetting ngram_mod\n", __func__, n_low);
+                    }
 
                     mod.reset();
                     n_low = 0;
@@ -756,6 +794,14 @@ struct common_speculative_state_ngram_mod : public common_speculative_state {
             }
         }
     }
+
+    int32_t n_max(const common_params_speculative & params) const override {
+        return params.ngram_mod.n_max;
+    }
+
+    int32_t n_min(const common_params_speculative & params) const override {
+        return params.ngram_mod.n_min;
+    }
 };
 
 struct common_speculative_state_ngram_cache : public common_speculative_state {
@@ -849,6 +895,14 @@ struct common_speculative_state_ngram_cache : public common_speculative_state {
         // TODO: noop
         GGML_UNUSED(n_accepted);
     }
+
+    int32_t n_max(const common_params_speculative & /*params*/) const override {
+        return n_draft;
+    }
+
+    int32_t n_min(const common_params_speculative & /*params*/) const override {
+        return 0;
+    }
 };
 
 struct common_speculative {
@@ -857,11 +911,13 @@ struct common_speculative {
     common_speculative_state * curr_impl = nullptr; // current implementation in use (for stats)
 };
 
-static common_ngram_map get_common_ngram_map(const common_speculative_config & config) {
-    uint16_t size_key   = config.params.ngram_size_n;
-    uint16_t size_value = config.params.ngram_size_m;
-    bool     key_only   = (config.type == COMMON_SPECULATIVE_TYPE_NGRAM_MAP_K);
-    uint16_t min_hits   = config.params.ngram_min_hits;
+static common_ngram_map get_common_ngram_map(
+        common_speculative_type type,
+        const common_params_speculative_ngram_map & config) {
+    uint16_t size_key   = config.size_n;
+    uint16_t size_value = config.size_m;
+    bool     key_only   = type == COMMON_SPECULATIVE_TYPE_NGRAM_MAP_K;
+    uint16_t min_hits   = config.min_hits;
 
     return common_ngram_map(size_key, size_value, key_only, min_hits);
 }
@@ -919,8 +975,8 @@ common_speculative * common_speculative_init(
         common_params_speculative & params,
         llama_context             * ctx_tgt) {
     llama_context * ctx_dft = nullptr;
-    if (params.model_dft) {
-        ctx_dft = llama_init_from_model(params.model_dft, params.cparams_dft);
+    if (params.draft.model) {
+        ctx_dft = llama_init_from_model(params.draft.model, params.draft.cparams);
         if (ctx_dft == nullptr) {
             LOG_ERR("%s", "failed to create draft context\n");
             return nullptr;
@@ -930,7 +986,7 @@ common_speculative * common_speculative_init(
     // Compute the implementations to use based on the config and their order of preference
     std::vector<common_speculative_config> configs = {}; // list of speculative configs to try
     {
-        bool has_draft = !params.mparams_dft.path.empty();
+        bool has_draft = !params.draft.mparams.path.empty();
         bool has_draft_eagle3 = false; // TODO PR-18039: if params.speculative.eagle3
 
         bool has_ngram_cache   = (params.type == COMMON_SPECULATIVE_TYPE_NGRAM_CACHE);
@@ -953,16 +1009,17 @@ common_speculative * common_speculative_init(
             configs.push_back(common_speculative_config(COMMON_SPECULATIVE_TYPE_NGRAM_MAP_K4V, params));
         }
         if (has_ngram_mod) {
-            // shared instance for all speculative decoding contexts
-            if (!params.ngram_mod) {
-                params.ngram_mod = std::make_shared<common_ngram_mod>(params.ngram_size_n, 4*1024*1024);
+            auto & sparams = params.ngram_mod;
 
-                LOG_INF("%s: initialized ngram_mod with n=%d, size=%zu (%.3f MB)\n", __func__,
-                        params.ngram_size_n, params.ngram_mod->size(),
-                        (float)(params.ngram_mod->size_bytes())/1024/1024);
+            if (!sparams.obj) {
+                sparams.obj = std::make_shared<common_ngram_mod>(sparams.n_match, 4*1024*1024);
 
-                if (params.ngram_size_n < 16) {
-                    LOG_WRN("%s: ngram_mod n=%d is too small - poor quality is possible, see: https://github.com/ggml-org/llama.cpp/pull/19164\n", __func__, params.ngram_size_n);
+                LOG_INF("%s: initialized ngram_mod with n_match=%d, size=%zu (%.3f MB)\n", __func__,
+                        sparams.n_match, sparams.obj->size(), (float)(sparams.obj->size_bytes())/1024/1024);
+
+                if (sparams.n_match < 16) {
+                    LOG_WRN("%s: ngram_mod n_match=%d is too small - poor quality is possible, "
+                            "see: https://github.com/ggml-org/llama.cpp/pull/19164\n", __func__, sparams.n_match);
                 }
             }
 
@@ -992,7 +1049,7 @@ common_speculative * common_speculative_init(
                 impls.push_back(std::make_unique<common_speculative_state_draft>(config.type,
                     /* .ctx_tgt      = */ ctx_tgt,
                     /* .ctx_dft      = */ ctx_dft,
-                    /* .replacements = */ params.replacements,
+                    /* .replacements = */ params.draft.replacements,
                     /* .use_ckpt     = */ use_ckpt
                 ));
                 break;
@@ -1002,18 +1059,18 @@ common_speculative * common_speculative_init(
                 break;
             }
             case COMMON_SPECULATIVE_TYPE_NGRAM_SIMPLE: {
-                common_ngram_map ngram_map = get_common_ngram_map(config);
+                common_ngram_map ngram_map = get_common_ngram_map(config.type, config.params.ngram_simple);
 
                 uint16_t ngram_size_key   = ngram_map.size_key;
                 uint16_t mgram_size_value = ngram_map.size_value;
 
                 auto config_simple = common_ngram_simple_config {
-                    /* .size_ngram      = */ ngram_size_key,
-                    /* .size_mgram      = */ mgram_size_value
+                    /* .size_ngram = */ ngram_size_key,
+                    /* .size_mgram = */ mgram_size_value
                 };
                 auto state = std::make_unique<common_speculative_state_ngram_simple>(
-                    /* .type            = */ config.type,
-                    /* .state           = */ config_simple
+                    /* .type  = */ config.type,
+                    /* .state = */ config_simple
                 );
                 impls.push_back(std::move(state));
                 break;
@@ -1022,18 +1079,17 @@ common_speculative * common_speculative_init(
             case COMMON_SPECULATIVE_TYPE_NGRAM_MAP_K4V: {
                 impls.push_back(std::make_unique<common_speculative_state_ngram_map_k>(
                     (config.type),
-                    get_common_ngram_map(config)
+                    get_common_ngram_map(config.type, config.params.ngram_map_k)
                 ));
                 break;
             }
             case COMMON_SPECULATIVE_TYPE_NGRAM_MOD: {
-                GGML_ASSERT(config.params.ngram_mod);
-                impls.push_back(std::make_unique<common_speculative_state_ngram_mod>(config.type, *config.params.ngram_mod));
+                GGML_ASSERT(config.params.ngram_mod.obj);
+                impls.push_back(std::make_unique<common_speculative_state_ngram_mod>(config.type, *config.params.ngram_mod.obj));
                 break;
             }
             case COMMON_SPECULATIVE_TYPE_NGRAM_CACHE: {
-                auto state = create_state_ngram_cache(
-                        params.lookup_cache_static, params.lookup_cache_dynamic, config);
+                auto state = create_state_ngram_cache(params.ngram_cache.lookup_cache_static, params.ngram_cache.lookup_cache_dynamic, config);
                 impls.push_back(std::make_unique<common_speculative_state_ngram_cache>(state));
                 break;
             }
@@ -1091,6 +1147,15 @@ llama_tokens common_speculative_draft(
             impl->n_call_draft++;
         }
 
+        {
+            const int n_min = impl->n_min(params);
+
+            if (!result.empty() && (int) result.size() < n_min) {
+                LOG_DBG("%s: ignoring small draft: %d < %d\n", __func__, (int) result.size(), n_min);
+                result.clear();
+            }
+        }
+
         if (!result.empty()) {
             LOG_DBG("%s: called impl %s, hist size = %zu, call_count = %zu, gen = %zu\n", __func__,
                     common_speculative_type_to_str(impl.get()->type).c_str(), prompt_tgt.size(),
@@ -1100,7 +1165,7 @@ llama_tokens common_speculative_draft(
             impl->n_gen_drafts++;
             impl->n_gen_tokens += result.size();
 
-            break; // We have a draft, so break out of the loop and return it.
+            break; // we have a draft, so break out of the loop and return it.
         }
     }
 
@@ -1128,6 +1193,32 @@ void common_speculative_accept(common_speculative * spec, uint16_t n_accepted) {
     }
 }
 
+int32_t common_speculative_n_max(const common_speculative * spec, const common_params_speculative & params) {
+    if (spec == nullptr) {
+        return 0;
+    }
+
+    int32_t n_max = 0;
+    for (const auto & impl : spec->impls) {
+        n_max = std::max(n_max, impl->n_max(params));
+    }
+
+    return n_max;
+}
+
+int32_t common_speculative_n_min(const common_speculative * spec, const common_params_speculative & params) {
+    if (spec == nullptr) {
+        return 0;
+    }
+
+    int32_t n_min = 0;
+    for (const auto & impl : spec->impls) {
+        n_min = std::max(n_min, impl->n_min(params));
+    }
+
+    return n_min;
+}
+
 void common_speculative_print_stats(const common_speculative * spec) {
     if (spec == nullptr) {
         return;

common/speculative.h

@@ -33,6 +33,9 @@ llama_tokens common_speculative_draft(
 // informs the speculative decoder that n_accepted tokens were accepted by the target model
 void common_speculative_accept(common_speculative * spec, uint16_t n_accepted);
 
+int32_t common_speculative_n_max(const common_speculative * spec, const common_params_speculative & params);
+int32_t common_speculative_n_min(const common_speculative * spec, const common_params_speculative & params);
+
 // print statistics about the speculative decoding
 void common_speculative_print_stats(const common_speculative * spec);
 

convert_hf_to_gguf.py

@@ -272,6 +272,22 @@ class ModelBase:
 
         return tensors
 
+    @staticmethod
+    def _scale_is_trivial(scale: Tensor) -> bool:
+        return scale.numel() <= 1 and abs(float(scale.float().sum()) - 1.0) < 1e-6
+
+    def _write_scale_tensor(self, scale_name: str, scale: Tensor):
+        if not self._scale_is_trivial(scale):
+            scale_f32 = scale.float().numpy().flatten()
+            logger.info(f"  + {scale_name} (per-tensor scale, shape [{scale_f32.size}])")
+            self.gguf_writer.add_tensor(scale_name, scale_f32)
+
+    def _write_scales_tensor(self, scale_name: str, scales: list[float]):
+        if not np.allclose(scales, 1.0, atol=1e-6):
+            scale_vals = np.array(scales, dtype=np.float32)
+            logger.info(f"  + {scale_name} (per-expert scale, shape [{len(scales)}])")
+            self.gguf_writer.add_tensor(scale_name, scale_vals)
+
     def dequant_model(self):
         # If all quantized tensors were already handled (e.g. pure NVFP4), skip
         if self._is_nvfp4 and not any(k.endswith((".weight_scale", ".weight_scale_inv")) for k in self.model_tensors):
@@ -494,7 +510,7 @@ class ModelBase:
                         s = self.model_tensors[name]
                         self.model_tensors[weight_name] = lambda w=w, s=s: dequant_simple(w(), s(), None)
                         tensors_to_remove.append(name)
-                    if name.endswith((".k_scale", ".v_scale")):
+                    if name.endswith((".input_scale", ".k_scale", ".v_scale")):
                         tensors_to_remove.append(name)
             elif quant_method is not None:
                 raise NotImplementedError(f"Quant method is not yet supported: {quant_method!r}")
@@ -602,10 +618,6 @@ class ModelBase:
         raw = np.concatenate([d_grouped, qs_grouped], axis=-1).reshape(out_features, n_super * 36)
         return raw, [out_features, n_super * 64]
 
-    @staticmethod
-    def _nvfp4_scale2_is_trivial(scale2: Tensor) -> bool:
-        return scale2.numel() <= 1 and abs(float(scale2.float().sum()) - 1.0) < 1e-6
-
     def _repack_nvfp4(self, name: str, weight: Tensor, scale: Tensor, scale2: Tensor, input_scale: Tensor):
         if "language_model." in name:
             name = name.replace("language_model.", "")
@@ -616,19 +628,8 @@ class ModelBase:
         logger.info(f"Repacked {new_name} with shape {shape} and quantization NVFP4")
         self.gguf_writer.add_tensor(new_name, raw, raw_dtype=gguf.GGMLQuantizationType.NVFP4)
 
-        # Emit per-tensor scale2 as a separate F32 tensor when non-trivial
-        if not self._nvfp4_scale2_is_trivial(scale2):
-            scale2_f32 = scale2.float().numpy().flatten()
-            scale_name = new_name.replace(".weight", ".scale")
-            logger.info(f"  + {scale_name} (per-tensor NVFP4 scale2, shape [{scale2_f32.size}])")
-            self.gguf_writer.add_tensor(scale_name, scale2_f32)
-
-        # Emit per-tensor input_scale as a separate F32 tensor when non-trivial
-        if not self._nvfp4_scale2_is_trivial(input_scale):
-            input_scale_f32 = input_scale.float().numpy().flatten()
-            input_scale_name = new_name.replace(".weight", ".input_scale")
-            logger.info(f"  + {input_scale_name} (per-tensor NVFP4 input_scale, shape [{input_scale_f32.size}])")
-            self.gguf_writer.add_tensor(input_scale_name, input_scale_f32)
+        self._write_scale_tensor(new_name.replace(".weight", ".scale"), scale2)
+        self._write_scale_tensor(new_name.replace(".weight", ".input_scale"), input_scale)
 
     def _generate_nvfp4_tensors(self):
         # Per-layer expert merging to avoid holding all experts in memory
@@ -719,24 +720,17 @@ class ModelBase:
         logger.info(f"Repacked {new_name} with shape [{len(experts)}, {shape[0]}, {shape[1]}] and quantization NVFP4")
         self.gguf_writer.add_tensor(new_name, merged, raw_dtype=gguf.GGMLQuantizationType.NVFP4)
 
-        # Emit per-expert scale2 tensor if any expert has non-trivial scale2
         scales.sort(key=lambda x: x[0])
-        scale_vals = np.array([s[1] for s in scales], dtype=np.float32)
-        if not np.allclose(scale_vals, 1.0, atol=1e-6):
-            scale_name = new_name.replace(".weight", ".scale")
-            logger.info(f"  + {scale_name} (per-expert NVFP4 scale2, shape [{len(scales)}])")
-            self.gguf_writer.add_tensor(scale_name, scale_vals)
+        self._write_scales_tensor(new_name.replace(".weight", ".scale"), [s[1] for s in scales])
 
-        # Emit per-expert input_scale tensor if any expert has non-trivial input_scale
         input_scales.sort(key=lambda x: x[0])
-        input_scale_vals = np.array([s[1] for s in input_scales], dtype=np.float32)
-        if not np.allclose(input_scale_vals, 1.0, atol=1e-6):
-            input_scale_name = new_name.replace(".weight", ".input_scale")
-            logger.info(f"  + {input_scale_name} (per-expert NVFP4 input_scale, shape [{len(input_scales)}])")
-            self.gguf_writer.add_tensor(input_scale_name, input_scale_vals)
+        self._write_scales_tensor(new_name.replace(".weight", ".input_scale"), [s[1] for s in input_scales])
 
         del experts, merged
 
+    def _needs_nvfp4_processing(self) -> bool:
+        return True
+
     def prepare_tensors(self):
         # detect NVFP4 quantization (ModelOpt format)
         quant_algo = (self.hparams.get("quantization_config") or {}).get("quant_algo")
@@ -767,7 +761,7 @@ class ModelBase:
         # NVFP4 weights are repacked and written directly to gguf_writer.
         # This must run before dequant_model so NVFP4 tensors are removed
         # from model_tensors, leaving only non-NVFP4 (e.g. FP8) for dequant.
-        if self._is_nvfp4:
+        if self._is_nvfp4 and self._needs_nvfp4_processing():
             self._generate_nvfp4_tensors()
 
         self.dequant_model()
@@ -2199,6 +2193,10 @@ class MmprojModel(ModelBase):
                 # merge configs
                 self.preprocessor_config = {**self.preprocessor_config, **cfg}
 
+    def _needs_nvfp4_processing(self) -> bool:
+        # nvfp4 quantization applies to the text model only.
+        return False
+
     def get_vision_config(self) -> dict[str, Any] | None:
         config_name = "vision_config" if not self.is_mistral_format else "vision_encoder"
         return self.global_config.get(config_name)
@@ -2891,6 +2889,20 @@ class LlamaModel(TextModel):
                 .swapaxes(1, 2)
                 .reshape(weights.shape))
 
+    def _repack_nvfp4(self, name: str, weight: Tensor, scale: Tensor, scale2: Tensor, input_scale: Tensor):
+        # Mirror the BF16 Q/K RoPE permutation site in modify_tensors; the NVFP4 path bypasses it.
+        if self.undo_permute:
+            n_head = self.find_hparam(["n_heads", "num_attention_heads"], optional=True)
+            n_kv_head = self.find_hparam(["n_kv_heads", "num_key_value_heads"], optional=True)
+            if n_head is not None:
+                if name.endswith("q_proj.weight"):
+                    weight = LlamaModel.permute(weight, n_head, n_head)
+                    scale  = LlamaModel.permute(scale, n_head, n_head)
+                elif name.endswith("k_proj.weight"):
+                    weight = LlamaModel.permute(weight, n_head, n_kv_head)
+                    scale  = LlamaModel.permute(scale, n_head, n_kv_head)
+        super()._repack_nvfp4(name, weight, scale, scale2, input_scale)
+
     _experts: list[dict[str, Tensor]] | None = None
 
     def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None) -> Iterable[tuple[str, Tensor]]:
@@ -4459,6 +4471,12 @@ class NemotronNanoV2VLModel(MmprojModel):
         }
         return vision_config
 
+    def dequant_model(self):
+        if self._is_nvfp4:
+            # Skip nvfp4 quantization for vision/audio model.
+            return
+        super().dequant_model()
+
     def set_gguf_parameters(self):
         if "image_mean" not in self.preprocessor_config:
             self.preprocessor_config["image_mean"] = [0.485, 0.456, 0.406]
@@ -4482,6 +4500,10 @@ class NemotronNanoV2VLModel(MmprojModel):
         if "input_conditioner" in name:
             return
 
+        # mtmd does not support video yet so skip tensors related to video.
+        if "radio_model.model.patch_generator.video_embedder" in name:
+            return
+
         # RADIO's pos_embed doesn't have .weight suffix, but clip.cpp expects it
         if "patch_generator.pos_embed" in name:
             if not name.endswith(".weight"):
@@ -6650,7 +6672,7 @@ class BertModel(TextModel):
 
         tokens: list[bytes] = [f"[PAD{i}]".encode("utf-8") for i in range(vocab_size)]
         scores: list[float] = [-10000.0] * vocab_size
-        toktypes: list[int] = [SentencePieceTokenTypes.UNUSED] * vocab_size  # ty: ignore[invalid-assignment]
+        toktypes: list[int] = [SentencePieceTokenTypes.UNUSED] * vocab_size
 
         if isinstance(tokenizer, SentencePieceProcessor):
             for token_id in range(tokenizer.vocab_size()):
@@ -10829,7 +10851,11 @@ class NemotronHModel(GraniteHybridModel):
         # uses self.model_arch to build the tensor name map, and all MoE-specific
         # mappings would be missed if it were called with the default non-MoE arch.
         hparams = ModelBase.load_hparams(args[0], self.is_mistral_format)
-        if "num_experts_per_tok" in hparams:
+        has_moe_params = (
+            "num_experts_per_tok" in hparams
+            or (isinstance(hparams.get("llm_config"), dict) and "num_experts_per_tok" in hparams["llm_config"])
+        )
+        if has_moe_params:
             self.model_arch = gguf.MODEL_ARCH.NEMOTRON_H_MOE
             self.is_moe = True
 
@@ -10976,6 +11002,11 @@ class NemotronHModel(GraniteHybridModel):
         if name.startswith(("vision_model.", "mlp1.")):
             return
 
+        if name.startswith(("sound_encoder.")):
+            return
+        if name.startswith(("sound_projection.")):
+            return
+
         # Strip language_model. prefix for VLM models (e.g., Nemotron Nano 12B v2 VL)
         if name.startswith("language_model."):
             name = name[len("language_model."):]
@@ -12685,11 +12716,12 @@ class MistralModel(LlamaModel):
     def set_mistral_config(gguf_writer: gguf.GGUFWriter, hparams: dict):
         if "yarn" in hparams:
             yarn_params = hparams["yarn"]
+            mscale_all_dim = 1.0 if not yarn_params["apply_scale"] else 0.0
             gguf_writer.add_rope_scaling_type(gguf.RopeScalingType.YARN)
             gguf_writer.add_rope_scaling_factor(yarn_params["factor"])
             gguf_writer.add_rope_scaling_yarn_beta_fast(yarn_params["beta"])
             gguf_writer.add_rope_scaling_yarn_beta_slow(yarn_params["alpha"])
-            gguf_writer.add_rope_scaling_yarn_log_mul(1.0) # mscale_all_dim
+            gguf_writer.add_rope_scaling_yarn_log_mul(mscale_all_dim)
             gguf_writer.add_rope_scaling_orig_ctx_len(yarn_params["original_max_position_embeddings"])
 
         if "llama_4_scaling" in hparams:
@@ -13215,17 +13247,18 @@ class LazyTorchTensor(gguf.LazyBase):
     }
 
     # only used when byteswapping data. Only correct size is needed
+    # TODO: uncomment uint64, uint32, and uint16, ref: https://github.com/pytorch/pytorch/issues/58734
     _dtype_byteswap_map: dict[torch.dtype, type] = {
         torch.float64: np.float64,
         torch.float32: np.float32,
         torch.bfloat16: np.float16,
         torch.float16: np.float16,
         torch.int64: np.int64,
-        torch.uint64: np.uint64,
+        # torch.uint64: np.uint64,
         torch.int32: np.int32,
-        torch.uint32: np.uint32,
+        # torch.uint32: np.uint32,
         torch.int16: np.int16,
-        torch.uint16: np.uint16,
+        # torch.uint16: np.uint16,
         torch.int8: np.int8,
         torch.uint8: np.uint8,
         torch.bool: np.uint8,

docs/backend/SYCL.md

@@ -51,6 +51,12 @@ The packages for FP32 and FP16 would have different accuracy and performance on
 
 ## News
 
+- 2026.04
+
+  - Optimize mul_mat by reorder feature for data type: Q4_K, Q5_K, Q_K, Q8_0.
+  - Fused MoE.
+  - Upgrate CI and built package for oneAPI 2025.3.3, support Ubuntu 24.04 built package.
+
 - 2026.03
   - Support Flash-Attention: less memory usage, performance impact depends on LLM.
 
@@ -349,6 +355,12 @@ Choose one of following methods to run.
 ./examples/sycl/test.sh
 ```
 
+- Run llama-server:
+
+```sh
+./examples/sycl/start-svr.sh -m PATH/MODEL_FILE
+```
+
 2. Command line
 Launch inference
 
@@ -637,10 +649,18 @@ Choose one of following methods to run.
 
 1. Script
 
+- Run test:
+
 ```
 examples\sycl\win-test.bat
 ```
 
+- Run llama-server:
+
+```
+examples\sycl\win-start-svr.bat -m PATH\MODEL_FILE
+```
+
 2. Command line
 
 Launch inference

docs/ops.md

@@ -26,7 +26,7 @@ Legend:
 |                            CLAMP | ❌ | ✅ | ✅ | ✅ | ✅ | 🟡 | 🟡 | 🟡 | ✅ | ❌ | ❌ |
 |                           CONCAT | ❌ | ✅ | ✅ | 🟡 | ✅ | 🟡 | ✅ | ✅ | ✅ | ❌ | ❌ |
 |                             CONT | ❌ | 🟡 | ✅ | ✅ | ✅ | 🟡 | 🟡 | ✅ | 🟡 | ❌ | ❌ |
-|                          CONV_2D | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ❌ | ✅ | ❌ | ❌ | ❌ |
+|                          CONV_2D | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ❌ | ✅ | ✅ | ❌ | ❌ |
 |                       CONV_2D_DW | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | ❌ |
 |                          CONV_3D | ❌ | ❌ | ✅ | ❌ | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ |
 |                CONV_TRANSPOSE_1D | ❌ | ✅ | ✅ | ✅ | ✅ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ |
@@ -60,7 +60,7 @@ Legend:
 |                       GROUP_NORM | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ |
 |                      HARDSIGMOID | ❌ | ✅ | ✅ | 🟡 | ✅ | ❌ | ✅ | 🟡 | ✅ | ❌ | ❌ |
 |                        HARDSWISH | ❌ | ✅ | ✅ | 🟡 | ✅ | ❌ | ✅ | 🟡 | ✅ | ❌ | ❌ |
-|                           IM2COL | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ |
+|                           IM2COL | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ |
 |                        IM2COL_3D | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | ❌ |
 |                          L2_NORM | ❌ | ✅ | ✅ | ✅ | ✅ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ |
 |                       LEAKY_RELU | ❌ | ✅ | ✅ | ✅ | 🟡 | ❌ | ✅ | 🟡 | ❌ | ❌ | ❌ |
@@ -105,7 +105,7 @@ Legend:
 |                              SQR | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | 🟡 | 🟡 | ✅ | ❌ | ❌ |
 |                             SQRT | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | 🟡 | 🟡 | ✅ | ❌ | ❌ |
 |                         SSM_CONV | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ |
-|                         SSM_SCAN | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ | 🟡 | ❌ | ❌ | ❌ |
+|                         SSM_SCAN | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ | 🟡 | ✅ | ❌ | ❌ |
 |                             STEP | ❌ | ✅ | ✅ | 🟡 | ✅ | ❌ | ✅ | 🟡 | ✅ | ❌ | ❌ |
 |                              SUB | ❌ | ✅ | ✅ | ✅ | 🟡 | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ |
 |                              SUM | ❌ | 🟡 | ✅ | 🟡 | 🟡 | ❌ | 🟡 | 🟡 | 🟡 | ❌ | ❌ |

docs/speculative.md

@@ -33,18 +33,18 @@ An example to use this approach can be the rewriting of source code by a LLM.
 This implementation looks for the last n-gram in history that matches the current n-gram and creates a draft using the m tokens following the matched n-gram. It is the simplest self-speculative approach with minimal overhead.
 
 ```
-llama-server [...] --spec-type ngram-simple --draft-max 64
+llama-server [...] --spec-type ngram-simple --spec-draft-n-max 64
 ```
 
 #### n-gram Map Key (`ngram-map-k`)
 
-This implementation looks for the current n-gram of size n (called the _key_) in the token history. If the key n-gram is followed by the same m tokens (called the _mgram_) multiple times, it creates a draft using these m tokens. This approach requires a minimum number of occurrences (argument `--spec-ngram-min-hits`, default is 1) before generating drafts.
+This implementation looks for the current n-gram of size n (called the _key_) in the token history. If the key n-gram is followed by the same m tokens (called the _mgram_) multiple times, it creates a draft using these m tokens. This approach requires a minimum number of occurrences (argument `--spec-ngram-map-k-min-hits`, default is 1) before generating drafts.
 
 The number of accepted tokens is stored for each used n-gram.
 
 **Example:**
 ```
-llama-server [...] --spec-type ngram-map-k --draft-max 64
+llama-server [...] --spec-type ngram-map-k --spec-draft-n-max 64
 ```
 
 #### n-gram Map Key-4-Values (`ngram-map-k4v`)
@@ -55,7 +55,7 @@ The number of accepted tokens is stored for each used n-gram.
 
 **Example:** Server options to be used if there are a lot of longer repetitions.
 ```
-llama-server [...] --spec-type ngram-map-k4v --spec-ngram-size-n 8 --spec-ngram-size-m 8 --spec-ngram-min-hits 2 --draft-max 64
+llama-server [...] --spec-type ngram-map-k4v --spec-ngram-map-k4v-size-n 8 --spec-ngram-map-k4v-size-m 8 --spec-ngram-map-k4v-min-hits 2 --spec-draft-n-max 64
 ```
 
 ### n-gram Mod (`ngram-mod`)
@@ -80,9 +80,9 @@ Currently, a single hash pool is shared across all server slots, so different re
 # notes:
 # - small `n` are not recommended
 # - MoEs require long drafts
-# - dense models: can reduce `--draft-min` and `--draft-max`
+# - dense models: can reduce `--spec-ngram-mod-n-min` and `--spec-ngram-mod-n-max`
 
-llama-server ... --spec-type ngram-mod --spec-ngram-size-n 24 --draft-min 48 --draft-max 64
+llama-server ... --spec-type ngram-mod --spec-ngram-mod-n-match 24 --spec-ngram-mod-n-min 48 --spec-ngram-mod-n-max 64
 ```
 
 Applications:
@@ -105,21 +105,90 @@ Example Video:
 
 If a draft model is combined with a draftless decoding the draftless decoding has higher precedence.
 
+### General Speculative Parameters
+
 ```
---draft, --draft-n, --draft-max N       number of tokens to draft for speculative decoding (default: 16)
-                                        (env: LLAMA_ARG_DRAFT_MAX)
---draft-min, --draft-n-min N            minimum number of draft tokens to use for speculative decoding
-                                        (default: 0)
-                                        (env: LLAMA_ARG_DRAFT_MIN)
-[...]
 --spec-type [none|ngram-cache|ngram-simple|ngram-map-k|ngram-map-k4v|ngram-mod]
                                         type of speculative decoding to use when no draft model is provided
                                         (default: none)
---spec-ngram-size-n N                   ngram size N for ngram-simple/ngram-map speculative decoding, length
-                                        of lookup n-gram (default: 12)
---spec-ngram-size-m N                   ngram size M for ngram-simple/ngram-map speculative decoding, length
-                                        of draft m-gram (default: 48)
---spec-ngram-min-hits N                 minimum hits for ngram-map speculative decoding (default: 1)
+                                        (env: LLAMA_ARG_SPEC_TYPE)
+--spec-default                          use default speculative decoding
+```
+
+### Draft Model Parameters
+
+```
+--spec-draft-model, -md, --model-draft  FNAME
+                                        draft model for speculative decoding (default: unused)
+                                        (env: LLAMA_ARG_SPEC_DRAFT_MODEL)
+--spec-draft-hf, -hfd, -hfrd, --hf-repo-draft  <user>/<model>[:quant]
+                                        HuggingFace repository for the draft model
+--spec-draft-n-max                      N
+                                        number of tokens to draft for speculative decoding (default: 16)
+                                        (env: LLAMA_ARG_SPEC_DRAFT_N_MAX)
+--spec-draft-n-min                      N
+                                        minimum number of draft tokens to use for speculative decoding (default: 0)
+                                        (env: LLAMA_ARG_SPEC_DRAFT_N_MIN)
+--spec-draft-p-split, --draft-p-split   P
+                                        speculative decoding split probability (default: 0.10)
+                                        (env: LLAMA_ARG_SPEC_DRAFT_P_SPLIT)
+--spec-draft-p-min, --draft-p-min       P
+                                        minimum speculative decoding probability (greedy) (default: 0.75)
+                                        (env: LLAMA_ARG_SPEC_DRAFT_P_MIN)
+--spec-draft-ctx-size, -cd, --ctx-size-draft  N
+                                        size of the prompt context for the draft model (default: 0, 0 = loaded from model)
+                                        (env: LLAMA_ARG_SPEC_DRAFT_CTX_SIZE)
+--spec-draft-ngl, -ngld, --gpu-layers-draft, --n-gpu-layers-draft  N
+                                        max. number of draft model layers to store in VRAM, either an exact number, 'auto', or 'all' (default: auto)
+                                        (env: LLAMA_ARG_N_GPU_LAYERS_DRAFT)
+--spec-draft-device, -devd, --device-draft  <dev1,dev2,..>
+                                        comma-separated list of devices to use for offloading the draft model
+--spec-draft-replace, --spec-replace    TARGET  DRAFT
+                                        translate the string in TARGET into DRAFT if the draft model and main model are not compatible
+```
+
+### n-gram Mod Parameters
+
+```
+--spec-ngram-mod-n-match                N
+                                        ngram-mod lookup length (default: 24)
+--spec-ngram-mod-n-min                  N
+                                        minimum number of ngram tokens to use for ngram-based speculative decoding (default: 48)
+--spec-ngram-mod-n-max                  N
+                                        maximum number of ngram tokens to use for ngram-based speculative decoding (default: 64)
+```
+
+### n-gram Simple Parameters
+
+```
+--spec-ngram-simple-size-n              N
+                                        ngram size N for ngram-simple speculative decoding, length of lookup n-gram (default: 12)
+--spec-ngram-simple-size-m              N
+                                        ngram size M for ngram-simple speculative decoding, length of draft m-gram (default: 48)
+--spec-ngram-simple-min-hits            N
+                                        minimum hits for ngram-simple speculative decoding (default: 1)
+```
+
+### n-gram Map Key Parameters
+
+```
+--spec-ngram-map-k-size-n               N
+                                        ngram size N for ngram-map-k speculative decoding, length of lookup n-gram (default: 12)
+--spec-ngram-map-k-size-m               N
+                                        ngram size M for ngram-map-k speculative decoding, length of draft m-gram (default: 48)
+--spec-ngram-map-k-min-hits             N
+                                        minimum hits for ngram-map-k speculative decoding (default: 1)
+```
+
+### n-gram Map Key-4-Values Parameters
+
+```
+--spec-ngram-map-k4v-size-n             N
+                                        ngram size N for ngram-map-k4v speculative decoding, length of lookup n-gram (default: 12)
+--spec-ngram-map-k4v-size-m             N
+                                        ngram size M for ngram-map-k4v speculative decoding, length of draft m-gram (default: 48)
+--spec-ngram-map-k4v-min-hits           N
+                                        minimum hits for ngram-map-k4v speculative decoding (default: 1)
 ```
 
 ### `--spec-type TYPE`
@@ -140,21 +209,40 @@ Specifies a type of speculative decoding without draft model.
 ./llama-server [...] --spec-type ngram-simple
 ```
 
-### `--spec-ngram-size-n N`
+### `--spec-ngram-*-size-n N`
 
 Sets the size N of the lookup n-gram for n-gram map based speculative decoding.
 The n-gram size N determines how many tokens in a row to look back when searching for matching patterns.
 
-### `--spec-ngram-size-m M`
+Each n-gram implementation has its own parameter:
+
+- `--spec-ngram-simple-size-n` for `ngram-simple`
+- `--spec-ngram-map-k-size-n` for `ngram-map-k`
+- `--spec-ngram-map-k4v-size-n` for `ngram-map-k4v`
+- `--spec-ngram-mod-n-match` for `ngram-mod`
+
+### `--spec-ngram-*-size-m M`
 
 Sets the size M of the draft m-gram for n-gram map based speculative decoding.
 The m-gram size determines how many tokens to draft when a match is found.
 Larger values can provide more speedup but may reduce acceptance rate.
 
-### `--spec-ngram-min-hits H`
+Each n-gram implementation has its own parameter:
+
+- `--spec-ngram-simple-size-m` for `ngram-simple`
+- `--spec-ngram-map-k-size-m` for `ngram-map-k`
+- `--spec-ngram-map-k4v-size-m` for `ngram-map-k4v`
+
+### `--spec-ngram-*-min-hits H`
 
 This option defines how often a key has to appear in the token history to be used as a draft (default is 1).
 
+Each n-gram implementation has its own parameter:
+
+- `--spec-ngram-simple-min-hits` for `ngram-simple`
+- `--spec-ngram-map-k-min-hits` for `ngram-map-k`
+- `--spec-ngram-map-k4v-min-hits` for `ngram-map-k4v`
+
 ## Statistics
 Each speculative decoding implementation prints statistics.
 
@@ -180,4 +268,3 @@ statistics ngram_map_k: #calls(b,g,a) = 6 1690 26, #gen drafts = 26, #acc drafts
 - `#gen tokens`: number of tokens generated by this implementation (including rejected tokens)
 - `#acc tokens`: number of tokens accepted by the main model
 - `dur(b,g,a): durations of begin (new prompt), generation and accumulation (process acceptance).
-

examples/debug/debug.cpp

@@ -202,10 +202,14 @@ static bool run(llama_context * ctx, const common_params & params) {
     print_tokenized_prompt(ctx, tokens, params.prompt);
 
     if (params.save_logits) {
-        output_data output {ctx, model, params};
-        std::filesystem::path model_path{params.model.path};
-        std::string model_name{model_path.stem().string()};
-        save_output_data(output, model_name, params.logits_output_dir);
+        try {
+            output_data output {ctx, model, params};
+            std::filesystem::path model_path{params.model.path};
+            std::string model_name{model_path.stem().string()};
+            save_output_data(output, model_name, params.logits_output_dir);
+        } catch (const std::exception & e) {
+            LOG_ERR("%s : error saving logits: %s\n", __func__, e.what());
+        }
     }
 
     return true;
@@ -223,7 +227,7 @@ int main(int argc, char ** argv) {
     llama_backend_init();
     llama_numa_init(params.numa);
 
-    std::optional<base_callback_data> cb_data;
+    std::optional<common_debug_cb_user_data> cb_data;
     if (!params.save_logits) {
         cb_data.emplace(params, params.tensor_filter);
     }

examples/eval-callback/eval-callback.cpp

@@ -3,7 +3,6 @@
 #include "debug.h"
 #include "log.h"
 #include "llama.h"
-#include "llama-cpp.h"
 
 #include <clocale>
 #include <string>
@@ -38,7 +37,7 @@ static bool run(llama_context * ctx, const common_params & params) {
 int main(int argc, char ** argv) {
     std::setlocale(LC_NUMERIC, "C");
 
-    base_callback_data cb_data;
+    common_debug_cb_user_data cb_data;
 
     common_params params;
 
@@ -53,7 +52,7 @@ int main(int argc, char ** argv) {
 
     // pass the callback to the backend scheduler
     // it will be executed for each node during the graph computation
-    params.cb_eval = common_debug_cb_eval<false>;
+    params.cb_eval = common_debug_cb_eval;
     params.cb_eval_user_data = &cb_data;
     params.warmup = false;
 

examples/gen-docs/gen-docs.cpp

@@ -73,12 +73,12 @@ static void write_help(std::ostringstream & ss, const md_file & md) {
     auto ctx_arg = common_params_parser_init(params, md.ex);
 
     std::vector<common_arg *> common_options;
-    std::vector<common_arg *> sparam_options;
+    std::vector<common_arg *> sampling_options;
     std::vector<common_arg *> specific_options;
     for (auto & opt : ctx_arg.options) {
         // in case multiple LLAMA_EXAMPLE_* are set, we prioritize the LLAMA_EXAMPLE_* matching current example
-        if (opt.is_sparam) {
-            sparam_options.push_back(&opt);
+        if (opt.is_sampling) {
+            sampling_options.push_back(&opt);
         } else if (opt.in_example(ctx_arg.ex)) {
             specific_options.push_back(&opt);
         } else {
@@ -93,7 +93,7 @@ static void write_help(std::ostringstream & ss, const md_file & md) {
     ss << "### Common params\n\n";
     write_table(ss, common_options);
     ss << "\n\n### Sampling params\n\n";
-    write_table(ss, sparam_options);
+    write_table(ss, sampling_options);
     ss << "\n\n### " << md.specific_section_header << "\n\n";
     write_table(ss, specific_options);
 

examples/lookup/lookup-create.cpp

@@ -37,9 +37,9 @@ int main(int argc, char ** argv){
 
     common_ngram_cache ngram_cache;
     common_ngram_cache_update(ngram_cache, LLAMA_NGRAM_STATIC, LLAMA_NGRAM_STATIC, inp, inp.size(), true);
-    fprintf(stderr, "%s: hashing done, writing file to %s\n", __func__, params.speculative.lookup_cache_static.c_str());
+    fprintf(stderr, "%s: hashing done, writing file to %s\n", __func__, params.speculative.ngram_cache.lookup_cache_static.c_str());
 
-    common_ngram_cache_save(ngram_cache, params.speculative.lookup_cache_static);
+    common_ngram_cache_save(ngram_cache, params.speculative.ngram_cache.lookup_cache_static);
 
     return 0;
 }

examples/lookup/lookup-stats.cpp

@@ -24,7 +24,7 @@ int main(int argc, char ** argv){
         return 1;
     }
 
-    const int n_draft = params.speculative.n_max;
+    const int n_draft = params.speculative.draft.n_max;
 
     // init llama.cpp
     llama_backend_init();
@@ -49,18 +49,18 @@ int main(int argc, char ** argv){
     {
         const int64_t t_start_draft_us = ggml_time_us();
 
-        if (!params.speculative.lookup_cache_static.empty()) {
+        if (!params.speculative.ngram_cache.lookup_cache_static.empty()) {
             try {
-                ngram_cache_static = common_ngram_cache_load(params.speculative.lookup_cache_static);
+                ngram_cache_static = common_ngram_cache_load(params.speculative.ngram_cache.lookup_cache_static);
             } catch (std::ifstream::failure const &) {
-                LOG_ERR("failed to open static lookup cache: %s", params.speculative.lookup_cache_static.c_str());
+                LOG_ERR("failed to open static lookup cache: %s", params.speculative.ngram_cache.lookup_cache_static.c_str());
                 exit(1);
             }
         }
 
-        if (!params.speculative.lookup_cache_dynamic.empty()) {
+        if (!params.speculative.ngram_cache.lookup_cache_dynamic.empty()) {
             try {
-                ngram_cache_dynamic = common_ngram_cache_load(params.speculative.lookup_cache_dynamic);
+                ngram_cache_dynamic = common_ngram_cache_load(params.speculative.ngram_cache.lookup_cache_dynamic);
             } catch (std::ifstream::failure const &) {} // if the file does not exist it will simply be created at the end of the program
         }
 

examples/lookup/lookup.cpp

@@ -25,7 +25,7 @@ int main(int argc, char ** argv){
     }
 
     // max. number of additional tokens to draft if match is found
-    const int n_draft = params.speculative.n_max;
+    const int n_draft = params.speculative.draft.n_max;
 
     // init llama.cpp
     llama_backend_init();
@@ -54,18 +54,18 @@ int main(int argc, char ** argv){
         const int64_t t_start_draft_us = ggml_time_us();
         common_ngram_cache_update(ngram_cache_context, LLAMA_NGRAM_MIN, LLAMA_NGRAM_MAX, inp, inp.size(), false);
 
-        if (!params.speculative.lookup_cache_static.empty()) {
+        if (!params.speculative.ngram_cache.lookup_cache_static.empty()) {
             try {
-                ngram_cache_static = common_ngram_cache_load(params.speculative.lookup_cache_static);
+                ngram_cache_static = common_ngram_cache_load(params.speculative.ngram_cache.lookup_cache_static);
             } catch (std::ifstream::failure const &) {
-                LOG_ERR("failed to open static lookup cache: %s", params.speculative.lookup_cache_static.c_str());
+                LOG_ERR("failed to open static lookup cache: %s", params.speculative.ngram_cache.lookup_cache_static.c_str());
                 exit(1);
             }
         }
 
-        if (!params.speculative.lookup_cache_dynamic.empty()) {
+        if (!params.speculative.ngram_cache.lookup_cache_dynamic.empty()) {
             try {
-                ngram_cache_dynamic = common_ngram_cache_load(params.speculative.lookup_cache_dynamic);
+                ngram_cache_dynamic = common_ngram_cache_load(params.speculative.ngram_cache.lookup_cache_dynamic);
             } catch (std::ifstream::failure const &) {} // if the file does not exist it will simply be created at the end of the program
         }
 
@@ -213,7 +213,7 @@ int main(int argc, char ** argv){
 
     // Update dynamic ngram cache with context ngram cache and save it to disk:
     common_ngram_cache_merge(ngram_cache_dynamic, ngram_cache_context);
-    common_ngram_cache_save(ngram_cache_dynamic, params.speculative.lookup_cache_dynamic);
+    common_ngram_cache_save(ngram_cache_dynamic, params.speculative.ngram_cache.lookup_cache_dynamic);
 
     LOG("\n\n");
 

examples/speculative-simple/speculative-simple.cpp

@@ -43,7 +43,7 @@ int main(int argc, char ** argv) {
         return 1;
     }
 
-    if (params.speculative.mparams_dft.path.empty()) {
+    if (params.speculative.draft.mparams.path.empty()) {
         LOG_ERR("%s: --model-draft is required\n", __func__);
         return 1;
     }
@@ -77,7 +77,7 @@ int main(int argc, char ** argv) {
 
     // TODO: simplify this logic
     {
-        const auto & params_spec = params.speculative;
+        const auto & params_spec = params.speculative.draft;
 
         auto params_dft = params;
 
@@ -85,15 +85,15 @@ int main(int argc, char ** argv) {
         params_dft.n_ctx        = params_spec.n_ctx;
         params_dft.n_batch      = llama_n_ctx_seq(ctx_tgt);
         params_dft.devices      = params_spec.devices;
-        params_dft.model        = params_spec.mparams_dft;
+        params_dft.model        = params_spec.mparams;
         params_dft.n_gpu_layers = params_spec.n_gpu_layers;
 
         if (params_spec.cpuparams.n_threads > 0) {
-            params_dft.cpuparams.n_threads       = params.speculative.cpuparams.n_threads;
-            params_dft.cpuparams_batch.n_threads = params.speculative.cpuparams_batch.n_threads;
+            params_dft.cpuparams.n_threads       = params.speculative.draft.cpuparams.n_threads;
+            params_dft.cpuparams_batch.n_threads = params.speculative.draft.cpuparams_batch.n_threads;
         }
 
-        params_dft.tensor_buft_overrides = params.speculative.tensor_buft_overrides;
+        params_dft.tensor_buft_overrides = params.speculative.draft.tensor_buft_overrides;
 
         auto mparams_dft = common_model_params_to_llama(params_dft);
 
@@ -103,8 +103,8 @@ int main(int argc, char ** argv) {
             return 1;
         }
 
-        params.speculative.model_dft = model_dft.get();
-        params.speculative.cparams_dft = common_context_params_to_llama(params_dft);
+        params.speculative.draft.model = model_dft.get();
+        params.speculative.draft.cparams = common_context_params_to_llama(params_dft);
     }
 
     // Tokenize the prompt
@@ -187,16 +187,6 @@ int main(int argc, char ** argv) {
             // generate a new draft
             draft = common_speculative_draft(spec, params_spec, prompt_tgt, id_last);
 
-            if ((int) draft.size() > params_spec.n_max) {
-                LOG_WRN("draft size %zu exceeds max %d, truncating\n", draft.size(), params_spec.n_max);
-                draft.resize(params_spec.n_max);
-            }
-
-            if ((int) draft.size() < params_spec.n_min) {
-                LOG_DBG("ignoring small draft: %zu < %d\n", draft.size(), params_spec.n_min);
-                draft.clear();
-            }
-
             // save the original draft size
             n_draft = draft.size();
 
@@ -220,19 +210,12 @@ int main(int argc, char ** argv) {
             }
         }
 
-        GGML_ASSERT(n_draft > 0);
-
         // always have a token to evaluate from before - id_last
         common_batch_clear(batch_tgt);
         common_batch_add  (batch_tgt, id_last, n_past++, { 0 }, true);
 
         // evaluate the target model on [id_last, draft0, draft1, ..., draftN-1]
         {
-            // do not waste time on small drafts
-            if (draft.size() < (size_t) params_spec.n_min) {
-                draft.clear();
-            }
-
             for (size_t i = 0; i < draft.size(); ++i) {
                 common_batch_add(batch_tgt, draft[i], n_past + i, { 0 }, true);
             }
@@ -340,7 +323,7 @@ int main(int argc, char ** argv) {
     LOG_INF("decoded %4d tokens in %8.3f seconds, speed: %8.3f t/s\n", n_predict, (t_dec_end - t_dec_start) / 1e6f, n_predict  / ((t_dec_end - t_dec_start) / 1e6f));
 
     LOG_INF("\n");
-    LOG_INF("n_draft   = %d\n", params_spec.n_max);
+    LOG_INF("n_draft   = %d\n", params_spec.draft.n_max);
     LOG_INF("n_predict = %d\n", n_predict);
     LOG_INF("n_drafted = %d\n", n_drafted);
     LOG_INF("n_accept  = %d\n", n_accept);

examples/speculative/speculative.cpp

@@ -49,7 +49,7 @@ int main(int argc, char ** argv) {
         return 1;
     }
 
-    if (params.speculative.mparams_dft.path.empty()) {
+    if (params.speculative.draft.mparams.path.empty()) {
         LOG_ERR("%s: --model-draft is required\n", __func__);
         return 1;
     }
@@ -58,7 +58,7 @@ int main(int argc, char ** argv) {
     const int n_seq_dft = params.n_parallel;
 
     // probability threshold for splitting a draft branch (only for n_seq_dft > 1)
-    const float p_draft_split = params.speculative.p_split;
+    const float p_draft_split = params.speculative.draft.p_split;
 
     std::default_random_engine rng(params.sampling.seed == LLAMA_DEFAULT_SEED ? std::random_device()() : params.sampling.seed);
     std::uniform_real_distribution<> u_dist;
@@ -80,15 +80,15 @@ int main(int argc, char ** argv) {
     ctx_tgt   = llama_init_tgt->context();
 
     // load the draft model
-    params.devices = params.speculative.devices;
-    params.model = params.speculative.mparams_dft;
-    params.n_gpu_layers = params.speculative.n_gpu_layers;
-    if (params.speculative.cpuparams.n_threads > 0) {
-        params.cpuparams.n_threads = params.speculative.cpuparams.n_threads;
+    params.devices = params.speculative.draft.devices;
+    params.model = params.speculative.draft.mparams;
+    params.n_gpu_layers = params.speculative.draft.n_gpu_layers;
+    if (params.speculative.draft.cpuparams.n_threads > 0) {
+        params.cpuparams.n_threads = params.speculative.draft.cpuparams.n_threads;
     }
 
-    params.cpuparams_batch.n_threads = params.speculative.cpuparams_batch.n_threads;
-    params.tensor_buft_overrides     = params.speculative.tensor_buft_overrides;
+    params.cpuparams_batch.n_threads = params.speculative.draft.cpuparams_batch.n_threads;
+    params.tensor_buft_overrides     = params.speculative.draft.tensor_buft_overrides;
 
     auto llama_init_dft = common_init_from_params(params);
 
@@ -110,13 +110,21 @@ int main(int argc, char ** argv) {
         return 1;
     }
 
-    if (
-        llama_vocab_get_add_bos(vocab_tgt) != llama_vocab_get_add_bos(vocab_dft) ||
-        llama_vocab_get_add_eos(vocab_tgt) != llama_vocab_get_add_eos(vocab_dft) ||
-        llama_vocab_bos(vocab_tgt) != llama_vocab_bos(vocab_dft) ||
-        llama_vocab_eos(vocab_tgt) != llama_vocab_eos(vocab_dft)
-    ) {
-        LOG_ERR("%s: draft model special tokens must match target model to use speculation\n", __func__);
+    if (llama_vocab_get_add_bos(vocab_tgt) != llama_vocab_get_add_bos(vocab_dft) ||
+        (llama_vocab_get_add_bos(vocab_tgt) && llama_vocab_bos(vocab_tgt) != llama_vocab_bos(vocab_dft))) {
+        LOG_ERR("%s: draft model bos tokens must match target model to use speculation. add: %d - %d, id: %d - %d)\n",
+                __func__,
+                llama_vocab_get_add_bos(vocab_tgt), llama_vocab_get_add_bos(vocab_dft),
+                llama_vocab_bos(vocab_tgt), llama_vocab_bos(vocab_dft));
+        return 1;
+    }
+
+    if (llama_vocab_get_add_eos(vocab_tgt) != llama_vocab_get_add_eos(vocab_dft) ||
+        (llama_vocab_get_add_eos(vocab_tgt) && llama_vocab_eos(vocab_tgt) != llama_vocab_eos(vocab_dft))) {
+        LOG_ERR("%s: draft model eos tokens must match target model to use speculation. add: %d - %d, id: %d - %d)\n",
+                __func__,
+                llama_vocab_get_add_eos(vocab_tgt), llama_vocab_get_add_eos(vocab_dft),
+                llama_vocab_eos(vocab_tgt), llama_vocab_eos(vocab_dft));
         return 1;
     }
 
@@ -137,11 +145,12 @@ int main(int argc, char ** argv) {
         for (int i = SPEC_VOCAB_CHECK_START_TOKEN_ID; i < std::min(n_vocab_tgt, n_vocab_dft); ++i) {
             const char * token_text_tgt = llama_vocab_get_text(vocab_tgt, i);
             const char * token_text_dft = llama_vocab_get_text(vocab_dft, i);
+
             if (std::strcmp(token_text_tgt, token_text_dft) != 0) {
                 LOG_ERR("%s: draft model vocab must match target model to use speculation but ", __func__);
                 LOG_ERR("token %d content differs - target '%s', draft '%s'\n", i,
-                        common_token_to_piece(ctx_tgt, i).c_str(),
-                        common_token_to_piece(ctx_dft, i).c_str());
+                        common_token_to_piece(vocab_tgt, i).c_str(),
+                        common_token_to_piece(vocab_dft, i).c_str());
                 return 1;
             }
         }
@@ -183,7 +192,7 @@ int main(int argc, char ** argv) {
     //GGML_ASSERT(n_vocab == llama_vocab_n_tokens(model_dft));
 
     // how many tokens to draft each time
-    int n_draft = params.speculative.n_max;
+    int n_draft = params.speculative.draft.n_max;
 
     int n_predict = 0;
     int n_drafted = 0;

examples/sycl/start-svr.sh

@@ -0,0 +1,124 @@
+#!/bin/bash
+
+#  MIT license
+#  Copyright (C) 2024 Intel Corporation
+#  SPDX-License-Identifier: MIT
+
+Help() {
+  cat << EOF
+Usage: $(basename "$0") [OPTIONS]
+
+This script processes files with specified options.
+
+Options:
+  -h, --help    Display this help message and exit.
+  -c, --context <value>    Set context length. Bigger need more memory.
+  -p, --promote <value>    Prompt to start generation with.
+  -m, --model   <value>    Full model file path.
+  -mg,--main-gpu <value>   Set main GPU ID (0 - n) for single GPU mode.
+  -sm,--split-mode <value> How to split the model across multiple GPUs, one of:
+                            - none: use one GPU only
+                            - layer (default): split layers and KV across GPUs
+                            - row: split rows across GPUs
+  -ngl,--n-gpu-layers <value>  Max. number of layers to store in VRAM (default: -1)
+  -lv,--log-verbosity <value>  Set the verbosity threshold. Messages with a higher verbosity will be
+                               ignored. Values:
+                                - 0: generic output
+                                - 1: error
+                                - 2: warning
+                                - 3: info
+                                - 4: debug
+
+
+EOF
+}
+
+BIN_FILE=./build/bin/llama-server
+SEED=0
+GPUS_SETTING=""
+
+MODEL_FILE=../models/Qwen3.5-4B-Q4_0.gguf
+NGL=99
+CONTEXT=4096
+GGML_SYCL_DEVICE=-1
+SPLIT_MODE=layer
+LOG_VERBOSE=3
+while [[ $# -gt 0 ]]; do
+    case "$1" in
+        -c|--context)
+            CONTEXT=$2
+            # Shift twice to consume both the option flag and its value
+            shift
+            shift
+            ;;
+        -m|--model)
+            MODEL_FILE="$2"
+            # Shift twice to consume both the option flag and its value
+            shift
+            shift
+            ;;
+        -mg|--main-gpu)
+            GGML_SYCL_DEVICE=$2
+            SPLIT_MODE=none
+            # Shift twice to consume both the option flag and its value
+            shift
+            shift
+            ;;
+        -sm|--split-mode)
+            SPLIT_MODE=$2
+            # Shift twice to consume both the option flag and its value
+            shift
+            shift
+            ;;
+        -ngl|--n-gpu-layers)
+            NGL=$2
+            # Shift twice to consume both the option flag and its value
+            shift
+            shift
+            ;;
+        -lv|--log-verbosity)
+            LOG_VERBOSE=$2
+            # Shift twice to consume both the option flag and its value
+            shift
+            shift
+            ;;
+        -h|--help)
+            Help
+            exit 0
+            ;;
+        *)
+            # Handle unknown options or stop processing options
+            echo "Invalid option: $1"
+            # Optional: exit script or shift to treat remaining as positional args
+            exit 1
+            ;;
+    esac
+done
+
+
+
+source /opt/intel/oneapi/setvars.sh
+
+#export GGML_SYCL_DEBUG=1
+
+#ZES_ENABLE_SYSMAN=1, Support to get free memory of GPU by sycl::aspect::ext_intel_free_memory. Recommended to use when --split-mode = layer.
+
+#support malloc device memory more than 4GB.
+export UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS=1
+echo "UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS=${UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS}"
+
+if [ $GGML_SYCL_DEVICE -ne -1 ]; then
+    echo "Use $GGML_SYCL_DEVICE as main GPU"
+    #use signle GPU only
+    GPUS_SETTING="-mg $GGML_SYCL_DEVICE -sm ${SPLIT_MODE}"
+    export ONEAPI_DEVICE_SELECTOR="level_zero:${$GGML_SYCL_DEVICE}"
+    echo "ONEAPI_DEVICE_SELECTOR=${ONEAPI_DEVICE_SELECTOR}"
+else
+    echo "Use all Intel GPUs, including iGPU & dGPU"
+    GPUS_SETTING="-sm ${SPLIT_MODE}"
+ fi
+
+echo "run cmd: ZES_ENABLE_SYSMAN=1 ${BIN_FILE} -m ${MODEL_FILE} -no-cnv -p "${INPUT_PROMPT}" -n 200 -e -ngl ${NGL} -s ${SEED} -c ${CONTEXT} ${GPUS_SETTING} -lv ${LOG_VERBOSE}  --mmap "
+ZES_ENABLE_SYSMAN=1 ${BIN_FILE} -m ${MODEL_FILE} -ngl ${NGL} -s ${SEED} -c ${CONTEXT} ${GPUS_SETTING} -lv ${LOG_VERBOSE} --mmap --host 0.0.0.0 --port 8000
+
+

examples/sycl/test.sh

@@ -38,7 +38,7 @@ SEED=0
 GPUS_SETTING=""
 
 INPUT_PROMPT="Building a website can be done in 10 simple steps:\nStep 1:"
-MODEL_FILE=models/llama-2-7b.Q4_0.gguf
+MODEL_FILE=../models/llama-2-7b.Q4_0.gguf
 NGL=99
 CONTEXT=4096
 GGML_SYCL_DEVICE=-1
@@ -122,9 +122,10 @@ if [ $GGML_SYCL_DEVICE -ne -1 ]; then
     export ONEAPI_DEVICE_SELECTOR="level_zero:${$GGML_SYCL_DEVICE}"
     echo "ONEAPI_DEVICE_SELECTOR=${ONEAPI_DEVICE_SELECTOR}"
 else
-   echo "Use all Intel GPUs, including iGPU & dGPU"
+    echo "Use all Intel GPUs, including iGPU & dGPU"
+    GPUS_SETTING="-sm ${SPLIT_MODE}"
  fi
 
-echo "run cmd: ZES_ENABLE_SYSMAN=1 ${BIN_FILE} -m ${MODEL_FILE} -no-cnv -p "${INPUT_PROMPT}" -n 400 -e -ngl ${NGL} -s ${SEED} -c ${CONTEXT} ${GPUS_SETTING} -lv ${LOG_VERBOSE}  --mmap "
-ZES_ENABLE_SYSMAN=1 ${BIN_FILE} -m ${MODEL_FILE} -no-cnv -p "${INPUT_PROMPT}" -n 400 -e -ngl ${NGL} -s ${SEED} -c ${CONTEXT} ${GPUS_SETTING} -lv ${LOG_VERBOSE} --mmap
+echo "run cmd: ZES_ENABLE_SYSMAN=1 ${BIN_FILE} -m ${MODEL_FILE} -no-cnv -p "${INPUT_PROMPT}" -n 200 -e -ngl ${NGL} -s ${SEED} -c ${CONTEXT} ${GPUS_SETTING} -lv ${LOG_VERBOSE}  --mmap "
+ZES_ENABLE_SYSMAN=1 ${BIN_FILE} -m ${MODEL_FILE} -no-cnv -p "${INPUT_PROMPT}" -n 200 -e -ngl ${NGL} -s ${SEED} -c ${CONTEXT} ${GPUS_SETTING} -lv ${LOG_VERBOSE} --mmap
 

examples/sycl/win-start-svr.bat

@@ -0,0 +1,179 @@
+::  MIT license
+::  Copyright (C) 2024 Intel Corporation
+::  SPDX-License-Identifier: MIT
+
+@echo off
+setlocal EnableExtensions EnableDelayedExpansion
+
+set "BIN_FILE=.\build\bin\llama-server.exe"
+set "SEED=0"
+set "GPUS_SETTING="
+
+set "MODEL_FILE=..\models\Qwen3.5-4B-Q4_0.gguf"
+set "NGL=99"
+set "CONTEXT=4096"
+set "GGML_SYCL_DEVICE=-1"
+set "SPLIT_MODE=layer"
+set "LOG_VERBOSE=3"
+
+if "%~1"=="" goto after_args
+
+:parse_args
+if "%~1"=="" goto after_args
+
+if /I "%~1"=="-c" (
+  if "%~2"=="" goto missing_value
+  set "CONTEXT=%~2"
+  shift
+  shift
+  goto parse_args
+)
+if /I "%~1"=="--context" (
+  if "%~2"=="" goto missing_value
+  set "CONTEXT=%~2"
+  shift
+  shift
+  goto parse_args
+)
+
+if /I "%~1"=="-m" (
+  if "%~2"=="" goto missing_value
+  set "MODEL_FILE=%~2"
+  shift
+  shift
+  goto parse_args
+)
+if /I "%~1"=="--model" (
+  if "%~2"=="" goto missing_value
+  set "MODEL_FILE=%~2"
+  shift
+  shift
+  goto parse_args
+)
+
+if /I "%~1"=="-mg" (
+  if "%~2"=="" goto missing_value
+  set "GGML_SYCL_DEVICE=%~2"
+  set "SPLIT_MODE=none"
+  shift
+  shift
+  goto parse_args
+)
+if /I "%~1"=="--main-gpu" (
+  if "%~2"=="" goto missing_value
+  set "GGML_SYCL_DEVICE=%~2"
+  set "SPLIT_MODE=none"
+  shift
+  shift
+  goto parse_args
+)
+
+if /I "%~1"=="-sm" (
+  if "%~2"=="" goto missing_value
+  set "SPLIT_MODE=%~2"
+  shift
+  shift
+  goto parse_args
+)
+if /I "%~1"=="--split-mode" (
+  if "%~2"=="" goto missing_value
+  set "SPLIT_MODE=%~2"
+  shift
+  shift
+  goto parse_args
+)
+
+if /I "%~1"=="-ngl" (
+  if "%~2"=="" goto missing_value
+  set "NGL=%~2"
+  shift
+  shift
+  goto parse_args
+)
+if /I "%~1"=="--n-gpu-layers" (
+  if "%~2"=="" goto missing_value
+  set "NGL=%~2"
+  shift
+  shift
+  goto parse_args
+)
+
+if /I "%~1"=="-lv" (
+  if "%~2"=="" goto missing_value
+  set "LOG_VERBOSE=%~2"
+  shift
+  shift
+  goto parse_args
+)
+if /I "%~1"=="--log-verbosity" (
+  if "%~2"=="" goto missing_value
+  set "LOG_VERBOSE=%~2"
+  shift
+  shift
+  goto parse_args
+)
+
+if /I "%~1"=="-h" goto help
+if /I "%~1"=="--help" goto help
+
+echo Invalid option: %~1
+exit /b 1
+
+:missing_value
+echo Missing value for option: %~1
+exit /b 1
+
+:help
+echo Usage: %~n0 [OPTIONS]
+echo.
+echo This script processes files with specified options.
+echo.
+echo Options:
+echo   -h, --help    Display this help message and exit.
+echo   -c, --context ^<value^>    Set context length. Bigger need more memory.
+echo   -m, --model   ^<value^>    Full model file path.
+echo   -mg,--main-gpu ^<value^>   Set main GPU ID (0 - n) for single GPU mode.
+echo   -sm,--split-mode ^<value^> How to split the model across multiple GPUs, one of:
+echo                             - none: use one GPU only
+echo                             - layer (default): split layers and KV across GPUs
+echo                             - row: split rows across GPUs
+echo   -ngl,--n-gpu-layers ^<value^>  Max. number of layers to store in VRAM (default: -1)
+echo   -lv,--log-verbosity ^<value^>  Set the verbosity threshold. Messages with a higher verbosity will be
+echo                                ignored. Values:
+echo                                 - 0: generic output
+echo                                 - 1: error
+echo                                 - 2: warning
+echo                                 - 3: info
+echo                                 - 4: debug
+exit /b 0
+
+:after_args
+
+REM In Windows CMD, source is not available; call oneAPI setvars if present.
+if exist "C:\Program Files (x86)\Intel\oneAPI\setvars.bat" (
+  call "C:\Program Files (x86)\Intel\oneAPI\setvars.bat" >nul
+) else (
+  echo Warning: oneAPI setvars.bat not found. Continuing without environment setup.
+)
+
+REM Support malloc device memory more than 4GB.
+set "UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS=1"
+echo UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS=%UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS%
+
+if not "%GGML_SYCL_DEVICE%"=="-1" (
+  echo Use %GGML_SYCL_DEVICE% as main GPU
+  REM Use single GPU only.
+  set "GPUS_SETTING=-mg %GGML_SYCL_DEVICE% -sm %SPLIT_MODE%"
+  set "ONEAPI_DEVICE_SELECTOR=level_zero:%GGML_SYCL_DEVICE%"
+  echo ONEAPI_DEVICE_SELECTOR=%ONEAPI_DEVICE_SELECTOR%
+) else (
+  echo Use all Intel GPUs, including iGPU ^& dGPU
+  set "GPUS_SETTING=-sm %SPLIT_MODE%"
+)
+
+echo run cmd: ZES_ENABLE_SYSMAN=1 %BIN_FILE% -m "%MODEL_FILE%" -ngl %NGL% -s %SEED% -c %CONTEXT% %GPUS_SETTING% -lv %LOG_VERBOSE% --mmap --host 0.0.0.0 --port 8000
+set "ZES_ENABLE_SYSMAN=1"
+%BIN_FILE% -m "%MODEL_FILE%" -ngl %NGL% -s %SEED% -c %CONTEXT% %GPUS_SETTING% -lv %LOG_VERBOSE% --mmap --host 0.0.0.0 --port 8000
+
+endlocal
+

examples/sycl/win-test.bat

@@ -2,10 +2,200 @@
 ::  Copyright (C) 2024 Intel Corporation
 ::  SPDX-License-Identifier: MIT
 
-set INPUT2="Building a website can be done in 10 simple steps:\nStep 1:"
-@call "C:\Program Files (x86)\Intel\oneAPI\setvars.bat" intel64 --force
 
-:: support malloc device memory more than 4GB.
-set UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS=1
-set LOAD_MODE="--mmap"
-.\build\bin\llama-completion.exe -m models\llama-2-7b.Q4_0.gguf -no-cnv -p %INPUT2% -n 400 -e -ngl 99 -s 0 %LOAD_MODE%
+@echo off
+setlocal EnableExtensions EnableDelayedExpansion
+
+REM MIT license
+REM Copyright (C) 2024 Intel Corporation
+REM SPDX-License-Identifier: MIT
+
+set "BIN_FILE=.\build\bin\llama-completion.exe"
+set "SEED=0"
+set "GPUS_SETTING="
+
+set "INPUT_PROMPT=Building a website can be done in 10 simple steps:^nStep 1:"
+set "MODEL_FILE=..\models\llama-2-7b.Q4_0.gguf"
+set "NGL=99"
+set "CONTEXT=4096"
+set "GGML_SYCL_DEVICE=-1"
+set "SPLIT_MODE=layer"
+set "LOG_VERBOSE=3"
+
+if "%~1"=="" goto after_args
+
+:parse_args
+if "%~1"=="" goto after_args
+
+if /I "%~1"=="-c" (
+  if "%~2"=="" goto missing_value
+  set "CONTEXT=%~2"
+  shift
+  shift
+  goto parse_args
+)
+if /I "%~1"=="--context" (
+  if "%~2"=="" goto missing_value
+  set "CONTEXT=%~2"
+  shift
+  shift
+  goto parse_args
+)
+
+if /I "%~1"=="-p" (
+  if "%~2"=="" goto missing_value
+  set "INPUT_PROMPT=%~2"
+  shift
+  shift
+  goto parse_args
+)
+if /I "%~1"=="--promote" (
+  if "%~2"=="" goto missing_value
+  set "INPUT_PROMPT=%~2"
+  shift
+  shift
+  goto parse_args
+)
+
+if /I "%~1"=="-m" (
+  if "%~2"=="" goto missing_value
+  set "MODEL_FILE=%~2"
+  shift
+  shift
+  goto parse_args
+)
+if /I "%~1"=="--model" (
+  if "%~2"=="" goto missing_value
+  set "MODEL_FILE=%~2"
+  shift
+  shift
+  goto parse_args
+)
+
+if /I "%~1"=="-mg" (
+  if "%~2"=="" goto missing_value
+  set "GGML_SYCL_DEVICE=%~2"
+  set "SPLIT_MODE=none"
+  shift
+  shift
+  goto parse_args
+)
+if /I "%~1"=="--main-gpu" (
+  if "%~2"=="" goto missing_value
+  set "GGML_SYCL_DEVICE=%~2"
+  set "SPLIT_MODE=none"
+  shift
+  shift
+  goto parse_args
+)
+
+if /I "%~1"=="-sm" (
+  if "%~2"=="" goto missing_value
+  set "SPLIT_MODE=%~2"
+  shift
+  shift
+  goto parse_args
+)
+if /I "%~1"=="--split-mode" (
+  if "%~2"=="" goto missing_value
+  set "SPLIT_MODE=%~2"
+  shift
+  shift
+  goto parse_args
+)
+
+if /I "%~1"=="-ngl" (
+  if "%~2"=="" goto missing_value
+  set "NGL=%~2"
+  shift
+  shift
+  goto parse_args
+)
+if /I "%~1"=="--n-gpu-layers" (
+  if "%~2"=="" goto missing_value
+  set "NGL=%~2"
+  shift
+  shift
+  goto parse_args
+)
+
+if /I "%~1"=="-lv" (
+  if "%~2"=="" goto missing_value
+  set "LOG_VERBOSE=%~2"
+  shift
+  shift
+  goto parse_args
+)
+if /I "%~1"=="--log-verbosity" (
+  if "%~2"=="" goto missing_value
+  set "LOG_VERBOSE=%~2"
+  shift
+  shift
+  goto parse_args
+)
+
+if /I "%~1"=="-h" goto help
+if /I "%~1"=="--help" goto help
+
+echo Invalid option: %~1
+exit /b 1
+
+:missing_value
+echo Missing value for option: %~1
+exit /b 1
+
+:help
+echo Usage: %~n0 [OPTIONS]
+echo.
+echo This script processes files with specified options.
+echo.
+echo Options:
+echo   -h, --help    Display this help message and exit.
+echo   -c, --context ^<value^>    Set context length. Bigger need more memory.
+echo   -p, --promote ^<value^>    Prompt to start generation with.
+echo   -m, --model   ^<value^>    Full model file path.
+echo   -mg,--main-gpu ^<value^>   Set main GPU ID (0 - n) for single GPU mode.
+echo   -sm,--split-mode ^<value^> How to split the model across multiple GPUs, one of:
+echo                             - none: use one GPU only
+echo                             - layer (default): split layers and KV across GPUs
+echo                             - row: split rows across GPUs
+echo   -ngl,--n-gpu-layers ^<value^>  Max. number of layers to store in VRAM (default: -1)
+echo   -lv,--log-verbosity ^<value^>  Set the verbosity threshold. Messages with a higher verbosity will be
+echo                                ignored. Values:
+echo                                 - 0: generic output
+echo                                 - 1: error
+echo                                 - 2: warning
+echo                                 - 3: info
+echo                                 - 4: debug
+exit /b 0
+
+:after_args
+
+REM In Windows CMD, source is not available; call oneAPI setvars if present.
+if exist "C:\Program Files (x86)\Intel\oneAPI\setvars.bat" (
+  call "C:\Program Files (x86)\Intel\oneAPI\setvars.bat" >nul
+) else (
+  echo Warning: oneAPI setvars.bat not found. Continuing without environment setup.
+)
+
+REM Support malloc device memory more than 4GB.
+set "UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS=1"
+echo UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS=%UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS%
+
+if not "%GGML_SYCL_DEVICE%"=="-1" (
+  echo Use %GGML_SYCL_DEVICE% as main GPU
+  REM Use single GPU only.
+  set "GPUS_SETTING=-mg %GGML_SYCL_DEVICE% -sm %SPLIT_MODE%"
+  set "ONEAPI_DEVICE_SELECTOR=level_zero:%GGML_SYCL_DEVICE%"
+  echo ONEAPI_DEVICE_SELECTOR=%ONEAPI_DEVICE_SELECTOR%
+) else (
+  echo Use all Intel GPUs, including iGPU ^& dGPU
+  set "GPUS_SETTING=-sm %SPLIT_MODE%"
+)
+
+echo run cmd: ZES_ENABLE_SYSMAN=1 %BIN_FILE% -m %MODEL_FILE% -no-cnv -p "%INPUT_PROMPT%" -n 200 -e -ngl %NGL% -s %SEED% -c %CONTEXT% %GPUS_SETTING% -lv %LOG_VERBOSE% --mmap
+set "ZES_ENABLE_SYSMAN=1"
+%BIN_FILE% -m "%MODEL_FILE%" -no-cnv -p "%INPUT_PROMPT%" -n 200 -e -ngl %NGL% -s %SEED% -c %CONTEXT% %GPUS_SETTING% -lv %LOG_VERBOSE% --mmap
+
+endlocal
+

ggml/CMakeLists.txt

@@ -5,7 +5,7 @@ project("ggml" C CXX ASM)
 ### GGML Version
 set(GGML_VERSION_MAJOR 0)
 set(GGML_VERSION_MINOR 10)
-set(GGML_VERSION_PATCH 0)
+set(GGML_VERSION_PATCH 2)
 set(GGML_VERSION_BASE "${GGML_VERSION_MAJOR}.${GGML_VERSION_MINOR}.${GGML_VERSION_PATCH}")
 
 list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/cmake/")

ggml/src/CMakeLists.txt

@@ -470,11 +470,10 @@ endforeach()
 
 target_link_libraries(ggml-base PRIVATE Threads::Threads)
 
-find_library(MATH_LIBRARY m)
-if (MATH_LIBRARY)
-    if (NOT WIN32 OR NOT DEFINED ENV{ONEAPI_ROOT})
-        target_link_libraries(ggml-base PRIVATE ${MATH_LIBRARY})
-    endif()
+if (DEFINED MATH_LIBRARY)
+    target_link_libraries(ggml-base PRIVATE ${MATH_LIBRARY})
+elseif (NOT WIN32 AND NOT DEFINED ENV{ONEAPI_ROOT})
+    target_link_libraries(ggml-base PRIVATE m)
 endif()
 
 if (CMAKE_SYSTEM_NAME MATCHES "Android")

ggml/src/ggml-backend-meta.cpp

@@ -1205,40 +1205,57 @@ static void ggml_backend_meta_buffer_set_tensor(ggml_backend_buffer_t buffer, gg
 
     if (split_state.n_segments != 1) {
         GGML_ASSERT(split_state.axis >= 0 && split_state.axis < GGML_MAX_DIMS);
-        GGML_ASSERT(offset == 0);
-        GGML_ASSERT(size == ggml_nbytes(tensor));
         GGML_ASSERT(tensor->ne[3] == 1);
+
         size_t offset_data = 0;
         std::vector<size_t> simple_offsets(n_bufs, 0);
         if (split_state.axis == GGML_BACKEND_SPLIT_AXIS_0) {
             GGML_ASSERT(tensor->ne[2] == 1);
+
+            const size_t row_stride = tensor->nb[1];
+            GGML_ASSERT(offset % row_stride == 0);
+            GGML_ASSERT(size   % row_stride == 0);
+            const int64_t r_start = offset / row_stride;
+            const int64_t r_count = size   / row_stride;
+            GGML_ASSERT(r_start + r_count <= tensor->ne[1]);
+
             const int64_t blck_size = ggml_blck_size(tensor->type);
             for (size_t s = 0; s < split_state.n_segments; s++) {
                 for (size_t j = 0; j < n_bufs; j++) {
                     ggml_tensor * simple_tensor = ggml_backend_meta_buffer_simple_tensor(tensor, j);
                     GGML_ASSERT(split_state.ne[s*n_bufs + j] % blck_size == 0);
                     const size_t nbytes = split_state.ne[s*n_bufs + j]/blck_size * tensor->nb[0];
-                    ggml_backend_tensor_set_2d(simple_tensor, (const char *) data + offset_data, simple_offsets[j], nbytes,
-                        tensor->ne[1], simple_tensor->nb[1], tensor->nb[1]);
+                    ggml_backend_tensor_set_2d(simple_tensor, (const char *) data + offset_data,
+                        simple_offsets[j] + r_start * simple_tensor->nb[1], nbytes,
+                        r_count, simple_tensor->nb[1], tensor->nb[1]);
                     offset_data       += nbytes;
                     simple_offsets[j] += nbytes;
                 }
             }
-            GGML_ASSERT(offset_data*tensor->ne[1] == size);
+            GGML_ASSERT(offset_data*r_count == size);
             return;
         }
         GGML_ASSERT(split_state.axis == GGML_BACKEND_SPLIT_AXIS_1);
+
+        const size_t row_stride = tensor->nb[2];
+        GGML_ASSERT(offset % row_stride == 0);
+        GGML_ASSERT(size   % row_stride == 0);
+        const int64_t r_start = offset / row_stride;
+        const int64_t r_count = size   / row_stride;
+        GGML_ASSERT(r_start + r_count <= tensor->ne[2]);
+
         for (size_t s = 0; s < split_state.n_segments; s++) {
             for (size_t j = 0; j < n_bufs; j++) {
                 ggml_tensor * simple_tensor = ggml_backend_meta_buffer_simple_tensor(tensor, j);
                 const size_t nbytes = split_state.ne[s*n_bufs + j] * tensor->nb[1];
-                ggml_backend_tensor_set_2d(simple_tensor, (const char *) data + offset_data, simple_offsets[j], nbytes,
-                    tensor->ne[2], simple_tensor->nb[2], tensor->nb[2]);
+                ggml_backend_tensor_set_2d(simple_tensor, (const char *) data + offset_data,
+                    simple_offsets[j] + r_start * simple_tensor->nb[2], nbytes,
+                    r_count, simple_tensor->nb[2], tensor->nb[2]);
                 offset_data       += nbytes;
                 simple_offsets[j] += nbytes;
             }
         }
-        GGML_ASSERT(offset_data*tensor->ne[2] == size);
+        GGML_ASSERT(offset_data*r_count == size);
         return;
     }
 
@@ -1295,40 +1312,57 @@ static void ggml_backend_meta_buffer_get_tensor(ggml_backend_buffer_t buffer, co
 
     if (split_state.n_segments != 1) {
         GGML_ASSERT(split_state.axis >= 0 && split_state.axis < GGML_MAX_DIMS);
-        GGML_ASSERT(offset == 0);
-        GGML_ASSERT(size == ggml_nbytes(tensor));
         GGML_ASSERT(tensor->ne[3] == 1);
+
         size_t offset_data = 0;
         std::vector<size_t> simple_offsets(n_bufs, 0);
         if (split_state.axis == GGML_BACKEND_SPLIT_AXIS_0) {
             GGML_ASSERT(tensor->ne[2] == 1);
+
+            const size_t row_stride = tensor->nb[1];
+            GGML_ASSERT(offset % row_stride == 0);
+            GGML_ASSERT(size   % row_stride == 0);
+            const int64_t r_start = offset / row_stride;
+            const int64_t r_count = size   / row_stride;
+            GGML_ASSERT(r_start + r_count <= tensor->ne[1]);
+
             const int64_t blck_size = ggml_blck_size(tensor->type);
             for (size_t s = 0; s < split_state.n_segments; s++) {
                 for (size_t j = 0; j < n_bufs; j++) {
                     const ggml_tensor * simple_tensor = ggml_backend_meta_buffer_simple_tensor(tensor, j);
                     GGML_ASSERT(split_state.ne[s*n_bufs + j] % blck_size == 0);
                     const size_t nbytes = split_state.ne[s*n_bufs + j]/blck_size * tensor->nb[0];
-                    ggml_backend_tensor_get_2d(simple_tensor, (char *) data + offset_data, simple_offsets[j], nbytes,
-                        tensor->ne[1], simple_tensor->nb[1], tensor->nb[1]);
+                    ggml_backend_tensor_get_2d(simple_tensor, (char *) data + offset_data,
+                        simple_offsets[j] + r_start * simple_tensor->nb[1], nbytes,
+                        r_count, simple_tensor->nb[1], tensor->nb[1]);
                     offset_data       += nbytes;
                     simple_offsets[j] += nbytes;
                 }
             }
-            GGML_ASSERT(offset_data*tensor->ne[1] == size);
+            GGML_ASSERT(offset_data*r_count == size);
             return;
         }
         GGML_ASSERT(split_state.axis == GGML_BACKEND_SPLIT_AXIS_1);
+
+        const size_t row_stride = tensor->nb[2];
+        GGML_ASSERT(offset % row_stride == 0);
+        GGML_ASSERT(size   % row_stride == 0);
+        const int64_t r_start = offset / row_stride;
+        const int64_t r_count = size   / row_stride;
+        GGML_ASSERT(r_start + r_count <= tensor->ne[2]);
+
         for (size_t s = 0; s < split_state.n_segments; s++) {
             for (size_t j = 0; j < n_bufs; j++) {
                 const ggml_tensor * simple_tensor = ggml_backend_meta_buffer_simple_tensor(tensor, j);
                 const size_t nbytes = split_state.ne[s*n_bufs + j] * tensor->nb[1];
-                ggml_backend_tensor_get_2d(simple_tensor, (char *) data + offset_data, simple_offsets[j], nbytes,
-                    tensor->ne[2], simple_tensor->nb[2], tensor->nb[2]);
+                ggml_backend_tensor_get_2d(simple_tensor, (char *) data + offset_data,
+                    simple_offsets[j] + r_start * simple_tensor->nb[2], nbytes,
+                    r_count, simple_tensor->nb[2], tensor->nb[2]);
                 offset_data       += nbytes;
                 simple_offsets[j] += nbytes;
             }
         }
-        GGML_ASSERT(offset_data*tensor->ne[2] == size);
+        GGML_ASSERT(offset_data*r_count == size);
         return;
     }
 
@@ -1792,7 +1826,24 @@ static enum ggml_status ggml_backend_meta_graph_compute(ggml_backend_t backend,
                     continue;
                 }
 
-                i = get_i_delayed(i);
+                const int i_delayed = get_i_delayed(i);
+
+                // If we can delay the AllReduce we need to consider the interaction with zero-sized tensor slices.
+                // A backend with such a slice would normally have valid data after participating in the AllReduce with a node that has
+                //     its compute flag disabled and thus gets its data zeroed out.
+                // If the AllReduce is delayed then the nodes until that point also need to have their compute flag disabled.
+                if (i_delayed > i) {
+                    for (size_t j = 0; j < n_backends; j++) {
+                        auto & bcj = backend_ctx->backend_configs[j];
+                        if ((bcj.nodes[i]->flags & GGML_TENSOR_FLAG_COMPUTE) == 0) {
+                            for (int ii = i + 1; ii <= i_delayed; ii++) {
+                                bcj.nodes[ii]->flags &= ~GGML_TENSOR_FLAG_COMPUTE;
+                            }
+                        }
+                    }
+                }
+
+                i = i_delayed;
 
                 for (size_t j = 0; j < n_backends; j++) {
                     auto & bcj = backend_ctx->backend_configs[j];
@@ -2049,8 +2100,8 @@ static const ggml_backend_i ggml_backend_meta_i = {
     /* .free                    = */ ggml_backend_meta_free,
     /* .set_tensor_async        = */ ggml_backend_meta_set_tensor_async,
     /* .get_tensor_async        = */ ggml_backend_meta_get_tensor_async,
-    /* .get_tensor_2d_async     = */ nullptr,
     /* .set_tensor_2d_async     = */ nullptr,
+    /* .get_tensor_2d_async     = */ nullptr,
     /* .cpy_tensor_async        = */ nullptr,
     /* .synchronize             = */ ggml_backend_meta_synchronize,
     /* .graph_plan_create       = */ nullptr,

ggml/src/ggml-backend-reg.cpp

@@ -181,6 +181,12 @@ struct ggml_backend_registry {
             return;
         }
 
+        for (auto & entry : backends) {
+            if (entry.reg == reg) {
+                return;
+            }
+        }
+
 #ifndef NDEBUG
         GGML_LOG_DEBUG("%s: registered backend %s (%zu devices)\n",
             __func__, ggml_backend_reg_name(reg), ggml_backend_reg_dev_count(reg));
@@ -192,6 +198,12 @@ struct ggml_backend_registry {
     }
 
     void register_device(ggml_backend_dev_t device) {
+        for (auto & dev : devices) {
+            if (dev == device) {
+                return;
+            }
+        }
+
 #ifndef NDEBUG
         GGML_LOG_DEBUG("%s: registered device %s (%s)\n", __func__, ggml_backend_dev_name(device), ggml_backend_dev_description(device));
 #endif

ggml/src/ggml-blas/ggml-blas.cpp

@@ -262,9 +262,9 @@ static struct ggml_backend_i blas_backend_i = {
     /* .get_name                = */ ggml_backend_blas_get_name,
     /* .free                    = */ ggml_backend_blas_free,
     /* .set_tensor_async        = */ NULL,
-    /* .get_tensor_2d_async     = */ NULL,
-    /* .set_tensor_2d_async     = */ NULL,
     /* .get_tensor_async        = */ NULL,
+    /* .set_tensor_2d_async     = */ NULL,
+    /* .get_tensor_2d_async     = */ NULL,
     /* .cpy_tensor_async        = */ NULL,
     /* .synchronize             = */ NULL,
     /* .graph_plan_create       = */ NULL,

ggml/src/ggml-cann/aclnn_ops.cpp

@@ -25,6 +25,7 @@
 #include "ggml-impl.h"
 #include "ggml.h"
 
+
 #include <aclnnop/aclnn_add.h>
 #include <aclnnop/aclnn_add_rms_norm.h>
 #include <aclnnop/aclnn_addcdiv.h>
@@ -45,7 +46,9 @@
 #include <aclnnop/aclnn_fused_infer_attention_score_v2.h>
 #include <aclnnop/aclnn_ger.h>
 #include <aclnnop/aclnn_group_norm.h>
+#include <aclnnop/aclnn_gather_v2.h>
 #include <aclnnop/aclnn_grouped_matmul_v3.h>
+#include <aclnnop/aclnn_scatter.h>
 #include <aclnnop/aclnn_gt_scalar.h>
 #include <aclnnop/aclnn_im2col.h>
 #include <aclnnop/aclnn_index_copy.h>
@@ -62,6 +65,7 @@
 #include <aclnnop/aclnn_permute.h>
 #include <aclnnop/aclnn_pow.h>
 #include <aclnnop/aclnn_pow_tensor_tensor.h>
+#include <aclnnop/aclnn_recurrent_gated_delta_rule.h>
 #include <aclnnop/aclnn_reduce_sum.h>
 #include <aclnnop/aclnn_reflection_pad1d.h>
 #include <aclnnop/aclnn_repeat.h>
@@ -69,11 +73,15 @@
 #include <aclnnop/aclnn_rms_norm.h>
 #include <aclnnop/aclnn_roll.h>
 #include <aclnnop/aclnn_softmax.h>
+#include <aclnnop/aclnn_softmax_cross_entropy_with_logits.h>
 #include <aclnnop/aclnn_sub.h>
 #include <aclnnop/aclnn_sum.h>
 #include <aclnnop/aclnn_threshold.h>
 #include <aclnnop/aclnn_tril.h>
+#include <aclnnop/aclnn_triangular_solve.h>
 #include <aclnnop/aclnn_triu.h>
+#include <aclnnop/aclnn_logical_not.h>
+#include <aclnnop/aclnn_masked_fill_scalar.h>
 #include <aclnnop/aclnn_upsample_nearest_2d.h>
 #include <aclnnop/aclnn_weight_quant_batch_matmul_v2.h>
 #include <aclnnop/aclnn_zero.h>
@@ -151,6 +159,107 @@ void ggml_cann_op_unary_gated(std::function<void(ggml_backend_cann_context &, ac
     GGML_CANN_CALL_ACLNN_OP(ctx, InplaceMul, acl_dst.get(), acl_src1.get());
 }
 
+// Fused SwiGLU using aclnnSwiGlu: splits input along innermost dim, applies
+// SiLU to left half, multiplies by right half.
+//
+// Falls back to the generic two-kernel path when src[1] != nullptr (two
+// independent halves) or swapped != 0 (reversed activation order), as
+// aclnnSwiGlu only handles the single interleaved tensor in standard order.
+//
+// CANN tiling for SwiGlu requires (storageShapeDim + viewDims) to be even.
+// aclCreateTensor always uses storageShapeDim=1, so viewDims must be odd.
+// We use a 3D view (1+3=4, even) to satisfy this constraint while preserving
+// correct split semantics along the innermost (ne[0]) dimension.
+void ggml_cann_swiglu(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
+    auto silu_fn = [](ggml_backend_cann_context & ctx, aclTensor * acl_src, aclTensor * acl_dst) {
+        GGML_CANN_CALL_ACLNN_OP(ctx, Silu, acl_src, acl_dst);
+    };
+
+    const int32_t swapped = ggml_get_op_params_i32(dst, 1);
+    if (dst->src[1] != nullptr || swapped != 0) {
+        ggml_cann_op_unary_gated(silu_fn, ctx, dst);
+        return;
+    }
+
+    // aclnnSwiGlu requires the split dim (src->ne[0]) to be even; fall back otherwise.
+    if (dst->src[0]->ne[0] % 2 != 0) {
+        ggml_cann_op_unary_gated(silu_fn, ctx, dst);
+        return;
+    }
+
+    ggml_tensor * src0 = dst->src[0];
+    size_t elem_size = ggml_element_size(src0);
+
+    // src0 GGML: [2*ne0, ne1, ne2, ne3] → 3D view [2*ne0, ne1, ne2*ne3]
+    // CANN reversed: [ne2*ne3, ne1, 2*ne0], split along CANN dim 2 (last).
+    int64_t ne0_x2   = src0->ne[0];
+    int64_t ne1      = src0->ne[1];
+    int64_t ne23     = src0->ne[2] * src0->ne[3];
+    int64_t src3d_ne[] = { ne0_x2, ne1, ne23 };
+    size_t  src3d_nb[] = { (size_t)src0->nb[0], (size_t)src0->nb[1], (size_t)src0->nb[2] };
+    acl_tensor_ptr acl_src = ggml_cann_create_tensor(src0->data, ggml_cann_type_mapping(src0->type),
+                                                     elem_size, src3d_ne, src3d_nb, 3);
+
+    // dst GGML: [ne0, ne1, ne2, ne3] → 3D view [ne0, ne1, ne2*ne3]
+    int64_t ne0      = dst->ne[0];
+    int64_t dst3d_ne[] = { ne0, ne1, ne23 };
+    size_t  dst3d_nb[] = { (size_t)dst->nb[0], (size_t)dst->nb[1], (size_t)dst->nb[2] };
+    acl_tensor_ptr acl_dst = ggml_cann_create_tensor(dst->data, ggml_cann_type_mapping(dst->type),
+                                                     elem_size, dst3d_ne, dst3d_nb, 3);
+
+    // CANN tensor [ne23, ne1, 2*ne0]: split along CANN dim 2 (last) = 2*ne0.
+    GGML_CANN_CALL_ACLNN_OP(ctx, SwiGlu, acl_src.get(), (int64_t)2, acl_dst.get());
+}
+
+// Fused GeGLU using aclnnGeGluV3: splits input along ne[0] (CANN last dim),
+// activates the LEFT half with GELU, multiplies by right half.
+// approximate: 0=tanh, 1=none(erf). activateLeft=true matches GGML convention.
+// outGelu is a required-but-discard output buffer.
+//
+// Falls back to the generic two-kernel path when src[1] != nullptr (two
+// independent halves) or swapped != 0 (reversed activation order), as
+// aclnnGeGluV3 only handles the single interleaved tensor in standard order.
+void ggml_cann_geglu(ggml_backend_cann_context & ctx, ggml_tensor * dst, int64_t approximate) {
+    auto gelu_fn = [](ggml_backend_cann_context & ctx, aclTensor * acl_src, aclTensor * acl_dst) {
+        GGML_CANN_CALL_ACLNN_OP(ctx, Gelu, acl_src, acl_dst);
+    };
+
+    const int32_t swapped = ggml_get_op_params_i32(dst, 1);
+    if (dst->src[1] != nullptr || swapped != 0) {
+        ggml_cann_op_unary_gated(gelu_fn, ctx, dst);
+        return;
+    }
+
+    // aclnnGeGluV3 requires the split dim (src->ne[0]) to be even; fall back otherwise.
+    if (dst->src[0]->ne[0] % 2 != 0) {
+        ggml_cann_op_unary_gated(gelu_fn, ctx, dst);
+        return;
+    }
+
+    ggml_tensor * src0 = dst->src[0];
+    acl_tensor_ptr acl_src = ggml_cann_create_tensor(src0);
+    acl_tensor_ptr acl_dst = ggml_cann_create_tensor(dst);
+
+    // Allocate a temporary buffer for the required outGelu output (same shape as dst).
+    // Build contiguous strides since the pool allocation is a fresh buffer.
+    size_t  elem_size    = ggml_element_size(dst);
+    int64_t ne[GGML_MAX_DIMS] = { dst->ne[0], dst->ne[1], dst->ne[2], dst->ne[3] };
+    size_t  nb[GGML_MAX_DIMS];
+    nb[0] = elem_size;
+    for (int i = 1; i < GGML_MAX_DIMS; i++) {
+        nb[i] = nb[i - 1] * ne[i - 1];
+    }
+    size_t gelu_out_size = nb[GGML_MAX_DIMS - 1] * ne[GGML_MAX_DIMS - 1];
+    ggml_cann_pool_alloc gelu_out_alloc(ctx.pool(), gelu_out_size);
+
+    acl_tensor_ptr acl_gelu_out = ggml_cann_create_tensor(
+        gelu_out_alloc.get(), ggml_cann_type_mapping(dst->type), elem_size, ne, nb, GGML_MAX_DIMS);
+    // V3 adds activateLeft param; true → Gelu(left)*right, matching GGML convention.
+    // GGML dim 0 → CANN last dim (index GGML_MAX_DIMS-1 = 3 for 4D tensor).
+    GGML_CANN_CALL_ACLNN_OP(ctx, GeGluV3, acl_src.get(), (int64_t)(GGML_MAX_DIMS - 1), approximate, true,
+                             acl_dst.get(), acl_gelu_out.get());
+}
+
 /**
  * @brief Repeats elements of a tensor along each dimension according to the
  * specified repeat array.
@@ -445,28 +554,33 @@ void ggml_cann_l2_norm(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
     ggml_cann_pool_alloc temp_buffer_allocator(ctx.pool(), n_bytes);
     void *               buffer = temp_buffer_allocator.get();
 
-    int64_t div_ne[] = { 1, src->ne[1], src->ne[2], src->ne[3] };
-    size_t  div_nb[GGML_MAX_DIMS];
-    div_nb[0] = sizeof(float);
+    int64_t norm_ne[] = { 1, src->ne[1], src->ne[2], src->ne[3] };
+    size_t  norm_nb[GGML_MAX_DIMS];
+    norm_nb[0] = sizeof(float);
     for (int i = 1; i < GGML_MAX_DIMS; ++i) {
-        div_nb[i] = div_nb[i - 1] * div_ne[i - 1];
+        norm_nb[i] = norm_nb[i - 1] * norm_ne[i - 1];
     }
-    acl_tensor_ptr acl_div = ggml_cann_create_tensor(buffer, ACL_FLOAT, type_size, div_ne, div_nb, GGML_MAX_DIMS);
+    acl_tensor_ptr acl_norm = ggml_cann_create_tensor(buffer, ACL_FLOAT, sizeof(float), norm_ne, norm_nb, GGML_MAX_DIMS);
 
     std::vector<int64_t> norm_dims  = { 3 };
     acl_int_array_ptr    dims_array = ggml_cann_create_int_array(norm_dims.data(), norm_dims.size());
 
     float          p_value  = 2.0f;
     acl_scalar_ptr p_scalar = ggml_cann_create_scalar(&p_value, aclDataType::ACL_FLOAT);
-    GGML_CANN_CALL_ACLNN_OP(ctx, Norm, acl_src.get(), p_scalar.get(), dims_array.get(), true, acl_div.get());
+    GGML_CANN_CALL_ACLNN_OP(ctx, Norm, acl_src.get(), p_scalar.get(), dims_array.get(), true, acl_norm.get());
 
-    // Clamp norm to at least eps: scale = 1/fmaxf(norm, eps)
-    acl_scalar_ptr acl_min = ggml_cann_create_scalar(&eps, aclDataType::ACL_FLOAT);
-    float          flt_max = FLT_MAX;
-    acl_scalar_ptr acl_max = ggml_cann_create_scalar(&flt_max, aclDataType::ACL_FLOAT);
-    GGML_CANN_CALL_ACLNN_OP(ctx, Clamp, acl_div.get(), acl_min.get(), acl_max.get(), acl_div.get());
+    ggml_cann_pool_alloc clamp_buffer_allocator(ctx.pool());
+    acl_tensor_ptr       acl_clamped;
 
-    GGML_CANN_CALL_ACLNN_OP(ctx, Div, acl_src.get(), acl_div.get(), acl_dst.get());
+    if (eps > 0.0f) {
+        void *         clamp_buf  = clamp_buffer_allocator.alloc(n_bytes);
+        acl_clamped               = ggml_cann_create_tensor(clamp_buf, ACL_FLOAT, sizeof(float), norm_ne, norm_nb, GGML_MAX_DIMS);
+        acl_scalar_ptr eps_scalar = ggml_cann_create_scalar(&eps, aclDataType::ACL_FLOAT);
+        GGML_CANN_CALL_ACLNN_OP(ctx, ClampMin, acl_norm.get(), eps_scalar.get(), acl_clamped.get());
+    }
+
+    aclTensor * acl_div_input = acl_clamped ? acl_clamped.get() : acl_norm.get();
+    GGML_CANN_CALL_ACLNN_OP(ctx, Div, acl_src.get(), acl_div_input, acl_dst.get());
 }
 
 void ggml_cann_cross_entropy_loss(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
@@ -482,56 +596,30 @@ void ggml_cann_cross_entropy_loss(ggml_backend_cann_context & ctx, ggml_tensor *
     logits_nb[1]              = logits_nb[0] * logits_ne[0];
     acl_tensor_ptr acl_logits = ggml_cann_create_tensor(src0->data, ACL_FLOAT, sizeof(float), logits_ne, logits_nb, 2);
 
-    size_t               log_softmax_type_size = sizeof(float);
-    int64_t              log_softmax_n_bytes   = nr * nc * log_softmax_type_size;
-    ggml_cann_pool_alloc log_softmax_allocator(ctx.pool(), log_softmax_n_bytes);
-    void *               log_softmax_buffer = log_softmax_allocator.get();
-
-    int64_t log_softmax_ne[] = { nc, nr };
-    size_t  log_softmax_nb[2];
-    log_softmax_nb[0]              = log_softmax_type_size;
-    log_softmax_nb[1]              = log_softmax_nb[0] * log_softmax_ne[0];
-    acl_tensor_ptr acl_log_softmax = ggml_cann_create_tensor(log_softmax_buffer, ACL_FLOAT, log_softmax_type_size,
-                                                             log_softmax_ne, log_softmax_nb, 2);
-
-    GGML_CANN_CALL_ACLNN_OP(ctx, LogSoftmax, acl_logits.get(), 1, acl_log_softmax.get());
-
     int64_t labels_ne[] = { nc, nr };
     size_t  labels_nb[2];
     labels_nb[0]              = ggml_type_size(src1->type);
     labels_nb[1]              = labels_nb[0] * labels_ne[0];
     acl_tensor_ptr acl_labels = ggml_cann_create_tensor(src1->data, ACL_FLOAT, sizeof(float), labels_ne, labels_nb, 2);
 
-    size_t               mul_type_size = sizeof(float);
-    int64_t              mul_n_bytes   = nr * nc * mul_type_size;
-    ggml_cann_pool_alloc mul_allocator(ctx.pool(), mul_n_bytes);
-    void *               mul_buffer = mul_allocator.get();
+    size_t               loss_per_sample_type_size = sizeof(float);
+    int64_t              loss_per_sample_n_bytes   = nr * loss_per_sample_type_size;
+    ggml_cann_pool_alloc loss_per_sample_allocator(ctx.pool(), loss_per_sample_n_bytes);
+    void *               loss_per_sample_buffer = loss_per_sample_allocator.get();
 
-    int64_t mul_ne[] = { nc, nr };
-    size_t  mul_nb[2];
-    mul_nb[0]                     = mul_type_size;
-    mul_nb[1]                     = mul_nb[0] * mul_ne[0];
-    acl_tensor_ptr acl_mul_result = ggml_cann_create_tensor(mul_buffer, ACL_FLOAT, mul_type_size, mul_ne, mul_nb, 2);
+    int64_t loss_per_sample_ne[] = { nr };
+    size_t  loss_per_sample_nb[1];
+    loss_per_sample_nb[0] = loss_per_sample_type_size;
+    acl_tensor_ptr acl_loss_per_sample = ggml_cann_create_tensor(
+        loss_per_sample_buffer, ACL_FLOAT, loss_per_sample_type_size, loss_per_sample_ne, loss_per_sample_nb, 1);
 
-    GGML_CANN_CALL_ACLNN_OP(ctx, Mul, acl_log_softmax.get(), acl_labels.get(), acl_mul_result.get());
+    size_t               backprop_n_bytes = nr * nc * sizeof(float);
+    ggml_cann_pool_alloc backprop_allocator(ctx.pool(), backprop_n_bytes);
+    void *               backprop_buffer = backprop_allocator.get();
+    acl_tensor_ptr acl_backprop = ggml_cann_create_tensor(backprop_buffer, ACL_FLOAT, sizeof(float), logits_ne, logits_nb, 2);
 
-    size_t               sum_per_sample_type_size = sizeof(float);
-    int64_t              sum_per_sample_n_bytes   = nr * sum_per_sample_type_size;
-    ggml_cann_pool_alloc sum_per_sample_allocator(ctx.pool(), sum_per_sample_n_bytes);
-    void *               sum_per_sample_buffer = sum_per_sample_allocator.get();
-
-    int64_t sum_per_sample_ne[] = { nr };
-    size_t  sum_per_sample_nb[1];
-    sum_per_sample_nb[0]              = sum_per_sample_type_size;
-    acl_tensor_ptr acl_sum_per_sample = ggml_cann_create_tensor(
-        sum_per_sample_buffer, ACL_FLOAT, sum_per_sample_type_size, sum_per_sample_ne, sum_per_sample_nb, 1);
-
-    std::vector<int64_t> sum_dims   = { 1 };
-    acl_int_array_ptr    dims_array = ggml_cann_create_int_array(sum_dims.data(), sum_dims.size());
-    bool                 keep_dims  = false;
-
-    GGML_CANN_CALL_ACLNN_OP(ctx, ReduceSum, acl_mul_result.get(), dims_array.get(), keep_dims, ACL_FLOAT,
-                            acl_sum_per_sample.get());
+    GGML_CANN_CALL_ACLNN_OP(ctx, SoftmaxCrossEntropyWithLogits, acl_logits.get(), acl_labels.get(),
+                            acl_loss_per_sample.get(), acl_backprop.get());
 
     size_t               total_sum_type_size = sizeof(float);
     int64_t              total_sum_n_bytes   = 1 * total_sum_type_size;
@@ -547,11 +635,12 @@ void ggml_cann_cross_entropy_loss(ggml_backend_cann_context & ctx, ggml_tensor *
 
     std::vector<int64_t> total_sum_dims    = { 0 };
     acl_int_array_ptr total_sum_dims_array = ggml_cann_create_int_array(total_sum_dims.data(), total_sum_dims.size());
+    bool              keep_dims            = false;
 
-    GGML_CANN_CALL_ACLNN_OP(ctx, ReduceSum, acl_sum_per_sample.get(), total_sum_dims_array.get(), keep_dims, ACL_FLOAT,
+    GGML_CANN_CALL_ACLNN_OP(ctx, ReduceSum, acl_loss_per_sample.get(), total_sum_dims_array.get(), keep_dims, ACL_FLOAT,
                             acl_total_sum.get());
 
-    float          value        = -1.0f / static_cast<float>(nr);
+    float          value        = 1.0f / static_cast<float>(nr);
     acl_scalar_ptr scale_factor = ggml_cann_create_scalar(&value, aclDataType::ACL_FLOAT);
     acl_tensor_ptr acl_dst =
         ggml_cann_create_tensor(dst->data, ACL_FLOAT, sizeof(float), total_sum_ne, total_sum_nb, 1);
@@ -589,6 +678,33 @@ void ggml_cann_group_norm(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
                             acl_mean_out.get(), acl_rstd_out.get());
 }
 
+void ggml_cann_set(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
+    ggml_tensor * src0 = dst->src[0];
+    ggml_tensor * src1 = dst->src[1];
+
+    size_t nb1     = ((int32_t *) dst->op_params)[0];
+    size_t nb2     = ((int32_t *) dst->op_params)[1];
+    size_t nb3     = ((int32_t *) dst->op_params)[2];
+    size_t offset  = ((int32_t *) dst->op_params)[3];
+    bool   inplace = (bool) ((int32_t *) dst->op_params)[4];
+
+    size_t param_nb[] = { ggml_element_size(src0), nb1, nb2, nb3 };
+
+    // Create a view of dst at the target offset with src1's dimensions
+    acl_tensor_ptr acl_dst  = ggml_cann_create_tensor(dst, src1->ne, param_nb, GGML_MAX_DIMS, ACL_FORMAT_ND, offset);
+    acl_tensor_ptr acl_src1 = ggml_cann_create_tensor(src1);
+
+    if (!inplace) {
+        // First copy src0 to dst entirely
+        size_t cpy_size = ggml_nbytes(dst);
+        ACL_CHECK(
+            aclrtMemcpyAsync(dst->data, cpy_size, src0->data, cpy_size, ACL_MEMCPY_DEVICE_TO_DEVICE, ctx.stream()));
+    }
+
+    // Copy src1 into the target region of dst
+    GGML_CANN_CALL_ACLNN_OP(ctx, InplaceCopy, acl_dst.get(), acl_src1.get());
+}
+
 void ggml_cann_acc(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
     ggml_tensor * src0 = dst->src[0];
     ggml_tensor * src1 = dst->src[1];
@@ -652,6 +768,113 @@ void ggml_cann_sum(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
     aclnn_reduce_sum(ctx, dst, reduce_dims, 4);
 }
 
+void ggml_cann_cumsum(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
+    ggml_tensor * src = dst->src[0];
+    acl_tensor_ptr acl_src = ggml_cann_create_tensor(src);
+    acl_tensor_ptr acl_dst = ggml_cann_create_tensor(dst);
+    // GGML cumsum operates along dim 0 (innermost / ne[0]).
+    // ggml_cann_create_tensor reverses dimensions to [ne3,ne2,ne1,ne0],
+    // so GGML dim 0 maps to CANN dim 3 (the last dim of the 4-D tensor).
+    GGML_CANN_CALL_ACLNN_OP(ctx, Cumsum, acl_src.get(), (int64_t)3,
+                            ggml_cann_type_mapping(dst->type), acl_dst.get());
+}
+
+void ggml_cann_solve_tri(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
+    ggml_tensor * src0 = dst->src[0];  // A: [N, N, B2, B3] lower triangular
+    ggml_tensor * src1 = dst->src[1];  // B: [K, N, B2, B3]
+
+    acl_tensor_ptr acl_a = ggml_cann_create_tensor(src0);
+    acl_tensor_ptr acl_b = ggml_cann_create_tensor(src1);
+    acl_tensor_ptr acl_x = ggml_cann_create_tensor(dst);
+
+    // mOut: triangular copy of A (required output), same shape as A.
+    const size_t a_bytes = ggml_nbytes(src0);
+    ggml_cann_pool_alloc m_alloc(ctx.pool(), a_bytes);
+    acl_tensor_ptr acl_m = ggml_cann_create_tensor(
+        m_alloc.get(), ggml_cann_type_mapping(src0->type),
+        ggml_type_size(src0->type), src0->ne, src0->nb, GGML_MAX_DIMS);
+
+    // Solve AX = B: upper=false (lower tri), transpose=false, unitriangular=false.
+    GGML_CANN_CALL_ACLNN_OP(ctx, TriangularSolve,
+        acl_b.get(), acl_a.get(), false, false, false,
+        acl_x.get(), acl_m.get());
+}
+
+void ggml_cann_diag(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
+    ggml_tensor * src = dst->src[0];
+
+    GGML_ASSERT(src->ne[1] == 1);
+
+    const int64_t N       = src->ne[0];
+    const int64_t n_batch = src->ne[2] * src->ne[3];
+    const size_t  nb_f32  = sizeof(float);
+
+    // Fill dst with zeros.
+    acl_tensor_ptr acl_dst = ggml_cann_create_tensor(dst);
+    {
+        float          zero = 0.0f;
+        acl_scalar_ptr acl_zero = ggml_cann_create_scalar(&zero, ACL_FLOAT);
+        GGML_CANN_CALL_ACLNN_OP(ctx, InplaceFillScalar, acl_dst.get(), acl_zero.get());
+    }
+
+    // Copy src vector onto the diagonal of dst via strided views.
+    // src viewed as [N, n_batch], contiguous strides.
+    int64_t ne_vec[2]      = { N, n_batch };
+    size_t  nb_src_vec[2]  = { nb_f32, N * nb_f32 };
+    // dst diagonal view: stride (N+1)*4 steps along the diagonal.
+    size_t  nb_dst_diag[2] = { (N + 1) * nb_f32, N * N * nb_f32 };
+
+    acl_tensor_ptr acl_src_vec  = ggml_cann_create_tensor(src->data, ACL_FLOAT, nb_f32, ne_vec, nb_src_vec, 2);
+    acl_tensor_ptr acl_dst_diag = ggml_cann_create_tensor(dst->data, ACL_FLOAT, nb_f32, ne_vec, nb_dst_diag, 2);
+
+    GGML_CANN_CALL_ACLNN_OP(ctx, InplaceCopy, acl_dst_diag.get(), acl_src_vec.get());
+}
+
+void ggml_cann_fill(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
+    float c = ggml_get_op_params_f32(dst, 0);
+
+    acl_tensor_ptr acl_dst = ggml_cann_create_tensor(dst);
+    acl_scalar_ptr acl_c   = ggml_cann_create_scalar(&c, ACL_FLOAT);
+    GGML_CANN_CALL_ACLNN_OP(ctx, InplaceFillScalar, acl_dst.get(), acl_c.get());
+}
+
+void ggml_cann_tri(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
+    ggml_tensor * src = dst->src[0];
+
+    const int64_t S       = src->ne[0];
+    const int64_t n_batch = src->ne[2] * src->ne[3];
+    const size_t  nb_f32  = sizeof(float);
+
+    int64_t ne3d[3] = { S, S, n_batch };
+    size_t  nb3d[3] = { nb_f32, S * nb_f32, S * S * nb_f32 };
+
+    const ggml_tri_type ttype = (ggml_tri_type) ggml_get_op_params_i32(dst, 0);
+
+    acl_tensor_ptr acl_src = ggml_cann_create_tensor(src->data, ACL_FLOAT, nb_f32, ne3d, nb3d, 3);
+    acl_tensor_ptr acl_dst = ggml_cann_create_tensor(dst->data, ACL_FLOAT, nb_f32, ne3d, nb3d, 3);
+
+    switch (ttype) {
+        case GGML_TRI_TYPE_LOWER:
+            // Tril(-1): preserve row > col (strict lower), zero upper + diagonal.
+            GGML_CANN_CALL_ACLNN_OP(ctx, Tril, acl_src.get(), (int64_t)-1, acl_dst.get());
+            break;
+        case GGML_TRI_TYPE_UPPER_DIAG:
+            // Triu(0): preserve row <= col (upper + diagonal), zero strict lower.
+            GGML_CANN_CALL_ACLNN_OP(ctx, Triu, acl_src.get(), (int64_t)0, acl_dst.get());
+            break;
+        case GGML_TRI_TYPE_UPPER:
+            // Triu(1): preserve row < col (strict upper), zero lower + diagonal.
+            GGML_CANN_CALL_ACLNN_OP(ctx, Triu, acl_src.get(), (int64_t)1, acl_dst.get());
+            break;
+        case GGML_TRI_TYPE_LOWER_DIAG:
+            // Tril(0): preserve row >= col (lower + diagonal), zero strict upper.
+            GGML_CANN_CALL_ACLNN_OP(ctx, Tril, acl_src.get(), (int64_t)0, acl_dst.get());
+            break;
+        default:
+            GGML_ABORT("unsupported tri type");
+    }
+}
+
 void ggml_cann_upsample_nearest2d(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
     ggml_tensor *  src     = dst->src[0];
     acl_tensor_ptr acl_src = ggml_cann_create_tensor(src, nullptr, nullptr, 0, ACL_FORMAT_NCHW);
@@ -1695,152 +1918,90 @@ void ggml_cann_softmax(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
     aclnn_softmax(ctx, softmax_tensor.get(), 3, acl_dst.get());
 }
 
-/**
- * @brief Performs index select operation on a 4D tensor using the CANN backend.
- *
- * This function applies the `IndexSelect` operation along a specific dimension
- * of the source tensor (`src_buffer`) using the indices from the index tensor (`index`).
- * It iterates over the last two dimensions of the source tensor, creates the corresponding
- * CANN tensors for the source, index, and output slices, and executes the `IndexSelect`
- * operation for each slice.
- *
- * @param ctx The context for CANN backend operations.
- * @param src_buffer The source buffer containing the 4D input tensor data.
- * @param src_ne The dimensions of the source tensor.
- * @param src_nb The strides (byte offsets) of the source tensor.
- * @param dst_buffer The destination buffer where the output tensor data will be written.
- * @param dst_ne The dimensions of the destination tensor.
- * @param dst_nb The strides (byte offsets) of the destination tensor.
- * @param index The index tensor specifying the indices to select from the source tensor.
- * @param type The data type of the source and destination tensors.
- */
-static void aclnn_index_select_4d(ggml_backend_cann_context & ctx,
-                                  void *                      src_buffer,
-                                  int64_t *                   src_ne,
-                                  size_t *                    src_nb,
-                                  void *                      dst_buffer,
-                                  int64_t *                   dst_ne,
-                                  size_t *                    dst_nb,
-                                  ggml_tensor *               index,
-                                  ggml_type                   type) {
-    for (int64_t i = 0; i < src_ne[3]; i++) {
-        for (int64_t j = 0; j < src_ne[2]; j++) {
-            // src
-            acl_tensor_ptr acl_src_tensor =
-                ggml_cann_create_tensor((char *) src_buffer + i * src_nb[3] + j * src_nb[2],
-                                        ggml_cann_type_mapping(type), ggml_type_size(type), src_ne, src_nb, 2);
-
-            // index
-            acl_tensor_ptr acl_index = ggml_cann_create_tensor(
-                (char *) index->data + (i % index->ne[2]) * index->nb[2] + (j % index->ne[1]) * index->nb[1],
-                ggml_cann_type_mapping(index->type), ggml_element_size(index), index->ne, index->nb, 1);
-
-            // out
-            acl_tensor_ptr acl_out =
-                ggml_cann_create_tensor((char *) dst_buffer + i * dst_nb[3] + j * dst_nb[2],
-                                        ggml_cann_type_mapping(type), ggml_type_size(type), dst_ne, dst_nb, 2);
-            GGML_CANN_CALL_ACLNN_OP(ctx, IndexSelect, acl_src_tensor.get(), 0, acl_index.get(), acl_out.get());
-        }
-    }
-}
-
-/**
- * @brief Performs inplace index copy operation on a 4D tensor using the CANN backend.
- *
- * This function applies the `IndexCopy` operation along a specific dimension of the
- * destination tensor (`dst_buffer`) by copying elements from the source tensor (`src_buffer`)
- * to positions specified by the index tensor (`index`).
- * It iterates over the last two dimensions of the tensors, creates the corresponding
- * CANN tensors for source, index, and destination slices, and performs the index copy
- * operation for each slice.
- *
- * @param ctx The context for CANN backend operations.
- * @param src_buffer The source buffer containing the 4D input tensor data to be copied.
- * @param src_ne The dimensions of the source tensor.
- * @param src_nb The strides (byte offsets) of the source tensor.
- * @param dst_buffer The destination buffer where values will be copied to.
- * @param dst_ne The dimensions of the destination tensor.
- * @param dst_nb The strides (byte offsets) of the destination tensor.
- * @param index The index tensor specifying target positions in the destination tensor.
- * @param type The data type of the source and destination tensors.
- */
-static void aclnn_index_copy_4d(ggml_backend_cann_context & ctx,
-                                void *                      src_buffer,
-                                int64_t *                   src_ne,
-                                size_t *                    src_nb,
-                                void *                      dst_buffer,
-                                int64_t *                   dst_ne,
-                                size_t *                    dst_nb,
-                                ggml_tensor *               index,
-                                ggml_type                   type) {
-    for (int64_t i = 0; i < src_ne[3]; i++) {
-        for (int64_t j = 0; j < src_ne[2]; j++) {
-            // src
-            acl_tensor_ptr acl_src_tensor =
-                ggml_cann_create_tensor((char *) src_buffer + i * src_nb[3] + j * src_nb[2],
-                                        ggml_cann_type_mapping(type), ggml_type_size(type), src_ne, src_nb, 2);
-
-            // index
-            acl_tensor_ptr acl_index = ggml_cann_create_tensor(
-                (char *) index->data + (i % index->ne[2]) * index->nb[2] + (j % index->ne[1]) * index->nb[1],
-                ggml_cann_type_mapping(index->type), ggml_element_size(index), index->ne, index->nb, 1);
-
-            // out
-            acl_tensor_ptr acl_out =
-                ggml_cann_create_tensor((char *) dst_buffer + i * dst_nb[3] + j * dst_nb[2],
-                                        ggml_cann_type_mapping(type), ggml_type_size(type), dst_ne, dst_nb, 2);
-            GGML_CANN_CALL_ACLNN_OP(ctx, InplaceIndexCopy, acl_out.get(), 0, acl_index.get(), acl_src_tensor.get());
-        }
-    }
-}
 
 void ggml_cann_get_rows(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
-    ggml_tensor * src0 = dst->src[0];  // src
+    ggml_tensor * src0 = dst->src[0];  // weight
     ggml_tensor * src1 = dst->src[1];  // index
 
     GGML_ASSERT(dst->type == GGML_TYPE_F32 || dst->type == GGML_TYPE_F16
                 || dst->type == GGML_TYPE_BF16);
 
+    // n_idx: number of row indices per (i2, i3) batch slice.
+    // ggml guarantees: src0->ne[2] == src1->ne[1], src0->ne[3] == src1->ne[2], src1->ne[3] == 1.
+    const int64_t n_idx = src1->ne[0];
+
+    // Gather all (i2, i3) batch slices from src into dst.
+    // ggml_cann_create_tensor reverses dims, so ACL sees [ne1, ne0].
+    // GatherV2 with dim=0 gathers along ACL dim-0 == ggml ne[1] (the vocabulary / row axis).
+    // nb: the 4 strides of the source buffer (nb[0..1] for the 2D slice shape,
+    //     nb[2..3] for computing per-batch-slice base pointer offsets).
+    auto gather_batched = [&](void * src_base, aclDataType acl_type, size_t type_size,
+                              const size_t * nb) {
+        int64_t src_ne[2]  = { src0->ne[0], src0->ne[1] };
+        size_t  src_nb_2d[2] = { nb[0], nb[1] };
+        int64_t dst_ne[2]  = { src0->ne[0], n_idx };
+        size_t  dst_nb_2d[2] = { dst->nb[0], dst->nb[1] };
+        int64_t idx_ne[1]  = { n_idx };
+        size_t  idx_nb[1]  = { (size_t)ggml_element_size(src1) };
+
+        for (int64_t i3 = 0; i3 < src0->ne[3]; i3++) {
+            for (int64_t i2 = 0; i2 < src0->ne[2]; i2++) {
+                acl_tensor_ptr acl_src = ggml_cann_create_tensor(
+                    (char *)src_base + i3 * nb[3] + i2 * nb[2],
+                    acl_type, type_size, src_ne, src_nb_2d, 2);
+                acl_tensor_ptr acl_idx = ggml_cann_create_tensor(
+                    (char *)src1->data + i3 * src1->nb[2] + i2 * src1->nb[1],
+                    ggml_cann_type_mapping(src1->type), (size_t)ggml_element_size(src1),
+                    idx_ne, idx_nb, 1);
+                acl_tensor_ptr acl_dst = ggml_cann_create_tensor(
+                    (char *)dst->data + i3 * dst->nb[3] + i2 * dst->nb[2],
+                    acl_type, type_size, dst_ne, dst_nb_2d, 2);
+                GGML_CANN_CALL_ACLNN_OP(ctx, GatherV2, acl_src.get(), 0, acl_idx.get(), acl_dst.get());
+            }
+        }
+    };
+
     switch (src0->type) {
         case GGML_TYPE_BF16:
         case GGML_TYPE_F16:
         case GGML_TYPE_F32:
             if (src0->type == dst->type) {
-                aclnn_index_select_4d(ctx, src0->data, src0->ne, src0->nb, dst->data, dst->ne, dst->nb, src1,
-                                      dst->type);
+                gather_batched(src0->data,
+                               ggml_cann_type_mapping(src0->type), ggml_type_size(src0->type),
+                               src0->nb);
             } else {
-                acl_tensor_ptr       acl_src0 = ggml_cann_create_tensor(src0);
-                ggml_cann_pool_alloc src_buffer_allocator(ctx.pool(), ggml_nelements(src0) * ggml_element_size(dst));
-                void *               src_trans_buffer = src_buffer_allocator.get();
-                size_t               src_trans_nb[GGML_MAX_DIMS];
-                src_trans_nb[0] = dst->nb[0];
+                // Cast src0 to dst type, then gather.
+                ggml_cann_pool_alloc src_cast_allocator(ctx.pool(),
+                                                        ggml_nelements(src0) * ggml_element_size(dst));
+                size_t src_cast_nb[GGML_MAX_DIMS];
+                src_cast_nb[0] = ggml_type_size(dst->type);
                 for (int i = 1; i < GGML_MAX_DIMS; i++) {
-                    src_trans_nb[i] = src_trans_nb[i - 1] * src0->ne[i - 1];
+                    src_cast_nb[i] = src_cast_nb[i - 1] * src0->ne[i - 1];
                 }
-                acl_tensor_ptr src_trans_tensor =
-                    ggml_cann_create_tensor(src_trans_buffer, ggml_cann_type_mapping(dst->type),
-                                            ggml_type_size(dst->type), src0->ne, src_trans_nb, GGML_MAX_DIMS);
-                aclnn_cast(ctx, acl_src0.get(), src_trans_tensor.get(), ggml_cann_type_mapping(dst->type));
-                aclnn_index_select_4d(ctx, src_trans_buffer, src0->ne, src_trans_nb, dst->data, dst->ne, dst->nb, src1,
-                                      dst->type);
+                acl_tensor_ptr acl_src0     = ggml_cann_create_tensor(src0);
+                acl_tensor_ptr acl_src_cast = ggml_cann_create_tensor(
+                    src_cast_allocator.get(), ggml_cann_type_mapping(dst->type), ggml_type_size(dst->type),
+                    src0->ne, src_cast_nb, GGML_MAX_DIMS);
+                aclnn_cast(ctx, acl_src0.get(), acl_src_cast.get(), ggml_cann_type_mapping(dst->type));
+
+                gather_batched(src_cast_allocator.get(),
+                               ggml_cann_type_mapping(dst->type), ggml_type_size(dst->type),
+                               src_cast_nb);
             }
             break;
         case GGML_TYPE_Q8_0:
             {
-                // add 1 dim for bcast mul.
+                // Dequantize Q8_0 to dst type, then gather.
                 size_t  weight_nb[GGML_MAX_DIMS + 1], scale_nb[GGML_MAX_DIMS + 1], dequant_nb[GGML_MAX_DIMS + 1];
                 int64_t weight_ne[GGML_MAX_DIMS + 1], scale_ne[GGML_MAX_DIMS + 1], *dequant_ne;
-                int64_t scale_offset = 0;
-                // [3,4,5,64] -> [3,4,5,2,32]
-                weight_ne[0]         = QK8_0;
-                weight_ne[1]         = src0->ne[0] / QK8_0;
-                weight_nb[0]         = sizeof(int8_t);
-                weight_nb[1]         = weight_nb[0] * weight_ne[0];
+                weight_ne[0] = QK8_0;
+                weight_ne[1] = src0->ne[0] / QK8_0;
+                weight_nb[0] = sizeof(int8_t);
+                weight_nb[1] = weight_nb[0] * weight_ne[0];
                 for (int i = 2; i < GGML_MAX_DIMS + 1; i++) {
                     weight_ne[i] = src0->ne[i - 1];
                     weight_nb[i] = weight_nb[i - 1] * weight_ne[i - 1];
                 }
-                // [3,4,5,64] -> [3,4,5,2,1]
                 scale_ne[0] = 1;
                 scale_ne[1] = src0->ne[0] / QK8_0;
                 scale_nb[0] = sizeof(uint16_t);
@@ -1849,31 +2010,33 @@ void ggml_cann_get_rows(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
                     scale_ne[i] = src0->ne[i - 1];
                     scale_nb[i] = scale_nb[i - 1] * scale_ne[i - 1];
                 }
-                // [3,4,5,64] -> [3,4,5,2,32]
                 dequant_ne    = weight_ne;
                 dequant_nb[0] = ggml_type_size(dst->type);
                 for (int i = 1; i < GGML_MAX_DIMS + 1; i++) {
                     dequant_nb[i] = dequant_nb[i - 1] * dequant_ne[i - 1];
                 }
-                scale_offset = ggml_nelements(src0) * sizeof(int8_t);
-                ggml_cann_pool_alloc dequant_buffer_allocator(ctx.pool(),
-                                                              ggml_nelements(src0) * ggml_type_size(dst->type));
-                acl_tensor_ptr       acl_weight_tensor = ggml_cann_create_tensor(src0->data, ACL_INT8, sizeof(int8_t),
-                                                                                 weight_ne, weight_nb, GGML_MAX_DIMS + 1);
-                acl_tensor_ptr       acl_scale_tensor =
-                    ggml_cann_create_tensor(src0->data, ACL_FLOAT16, sizeof(uint16_t), scale_ne, scale_nb,
-                                            GGML_MAX_DIMS + 1, ACL_FORMAT_ND, scale_offset);
-                acl_tensor_ptr dequant_tensor =
-                    ggml_cann_create_tensor(dequant_buffer_allocator.get(), ggml_cann_type_mapping(dst->type),
-                                            ggml_type_size(dst->type), dequant_ne, dequant_nb, GGML_MAX_DIMS + 1);
-                aclnn_mul(ctx, acl_weight_tensor.get(), acl_scale_tensor.get(), dequant_tensor.get());
-                dequant_nb[0] = ggml_type_size(dst->type);
+                const int64_t scale_offset = ggml_nelements(src0) * sizeof(int8_t);
+                ggml_cann_pool_alloc dequant_allocator(ctx.pool(),
+                                                       ggml_nelements(src0) * ggml_type_size(dst->type));
+                acl_tensor_ptr acl_weight = ggml_cann_create_tensor(src0->data, ACL_INT8, sizeof(int8_t),
+                                                                     weight_ne, weight_nb, GGML_MAX_DIMS + 1);
+                acl_tensor_ptr acl_scale  = ggml_cann_create_tensor(
+                    src0->data, ACL_FLOAT16, sizeof(uint16_t), scale_ne, scale_nb,
+                    GGML_MAX_DIMS + 1, ACL_FORMAT_ND, scale_offset);
+                acl_tensor_ptr acl_dequant = ggml_cann_create_tensor(
+                    dequant_allocator.get(), ggml_cann_type_mapping(dst->type),
+                    ggml_type_size(dst->type), dequant_ne, dequant_nb, GGML_MAX_DIMS + 1);
+                aclnn_mul(ctx, acl_weight.get(), acl_scale.get(), acl_dequant.get());
+
+                // Reinterpret dequant buffer as 4D [src0->ne] with contiguous strides.
                 dequant_ne    = src0->ne;
+                dequant_nb[0] = ggml_type_size(dst->type);
                 for (int i = 1; i < GGML_MAX_DIMS; i++) {
                     dequant_nb[i] = dequant_nb[i - 1] * src0->ne[i - 1];
                 }
-                aclnn_index_select_4d(ctx, dequant_buffer_allocator.get(), dequant_ne, dequant_nb, dst->data, dst->ne,
-                                      dst->nb, src1, dst->type);
+                gather_batched(dequant_allocator.get(),
+                               ggml_cann_type_mapping(dst->type), ggml_type_size(dst->type),
+                               dequant_nb);
                 break;
             }
         default:
@@ -1883,31 +2046,70 @@ void ggml_cann_get_rows(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
 }
 
 void ggml_cann_set_rows(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
-    ggml_tensor * src0 = dst->src[0];  // src
-    ggml_tensor * src1 = dst->src[1];  // index
+    ggml_tensor * src0 = dst->src[0];  // source values
+    ggml_tensor * src1 = dst->src[1];  // row indices
+
+    // n_idx: number of source rows to scatter per batch slice.
+    // ggml guarantees: src0->ne[1] == src1->ne[0].
+    const int64_t n_idx = src1->ne[0];
+
+    // Copy n_idx rows of src [ne0, n_idx] into dst [ne0, ne1] at positions given by a 1D index.
+    // ggml_cann_create_tensor reverses dims, so ACL sees [ne1, ne0] for dst.
+    // InplaceIndexCopy with dim=0 copies along ACL dim-0 == ggml ne[1] (the row axis).
+    // src_nb: the 4 strides of the source buffer (nb[0..1] for the 2D slice shape,
+    //         nb[2..3] for computing per-batch-slice base pointer offsets).
+    auto scatter_batched = [&](void * src_base, aclDataType acl_type, size_t type_size,
+                               const size_t * src_nb) {
+        int64_t d_ne[2]    = { dst->ne[0], dst->ne[1] };
+        size_t  d_nb[2]    = { dst->nb[0], dst->nb[1] };
+        int64_t s_ne[2]    = { dst->ne[0], n_idx };
+        size_t  s_nb_2d[2] = { src_nb[0], src_nb[1] };
+        int64_t i_ne[1]    = { n_idx };
+        size_t  i_nb[1]    = { (size_t)ggml_element_size(src1) };
+
+        for (int64_t i3 = 0; i3 < dst->ne[3]; i3++) {
+            for (int64_t i2 = 0; i2 < dst->ne[2]; i2++) {
+                acl_tensor_ptr acl_dst = ggml_cann_create_tensor(
+                    (char *)dst->data + i3 * dst->nb[3] + i2 * dst->nb[2],
+                    acl_type, type_size, d_ne, d_nb, 2);
+                acl_tensor_ptr acl_idx = ggml_cann_create_tensor(
+                    (char *)src1->data + (i3 % src1->ne[2]) * src1->nb[2] + (i2 % src1->ne[1]) * src1->nb[1],
+                    ggml_cann_type_mapping(src1->type), (size_t)ggml_element_size(src1),
+                    i_ne, i_nb, 1);
+                acl_tensor_ptr acl_src = ggml_cann_create_tensor(
+                    (char *)src_base + i3 * src_nb[3] + i2 * src_nb[2],
+                    acl_type, type_size, s_ne, s_nb_2d, 2);
+                GGML_CANN_CALL_ACLNN_OP(ctx, InplaceIndexCopy, acl_dst.get(), 0, acl_idx.get(), acl_src.get());
+            }
+        }
+    };
 
     switch (dst->type) {
         case GGML_TYPE_F32:
-            {
-                aclnn_index_copy_4d(ctx, src0->data, src0->ne, src0->nb, dst->data, dst->ne, dst->nb, src1, dst->type);
-                break;
-            }
+            scatter_batched(src0->data,
+                            ggml_cann_type_mapping(dst->type), ggml_type_size(dst->type),
+                            src0->nb);
+            break;
         case GGML_TYPE_F16:
         case GGML_TYPE_BF16:
             {
-                acl_tensor_ptr       acl_src0 = ggml_cann_create_tensor(src0);
-                ggml_cann_pool_alloc src_buffer_allocator(ctx.pool(), ggml_nelements(src0) * sizeof(uint16_t));
-                void *               src_trans_buffer = src_buffer_allocator.get();
-                size_t               src_trans_nb[GGML_MAX_DIMS];
-                src_trans_nb[0] = sizeof(uint16_t);
+                // Cast src0 (F32) to dst type first.
+                ggml_cann_pool_alloc src_cast_allocator(ctx.pool(),
+                                                        ggml_nelements(src0) * ggml_type_size(dst->type));
+                size_t src_cast_nb[GGML_MAX_DIMS];
+                src_cast_nb[0] = ggml_type_size(dst->type);
                 for (int i = 1; i < GGML_MAX_DIMS; i++) {
-                    src_trans_nb[i] = src_trans_nb[i - 1] * src0->ne[i - 1];
+                    src_cast_nb[i] = src_cast_nb[i - 1] * src0->ne[i - 1];
                 }
-                acl_tensor_ptr src_trans_tensor = ggml_cann_create_tensor(
-                    src_trans_buffer, ggml_cann_type_mapping(dst->type), ggml_type_size(dst->type), src0->ne, src_trans_nb, GGML_MAX_DIMS);
-                aclnn_cast(ctx, acl_src0.get(), src_trans_tensor.get(), ggml_cann_type_mapping(dst->type));
-                aclnn_index_copy_4d(ctx, src_trans_buffer, src0->ne, src_trans_nb, dst->data, dst->ne, dst->nb, src1,
-                                    dst->type);
+                acl_tensor_ptr acl_src0     = ggml_cann_create_tensor(src0);
+                acl_tensor_ptr acl_src_cast = ggml_cann_create_tensor(
+                    src_cast_allocator.get(), ggml_cann_type_mapping(dst->type), ggml_type_size(dst->type),
+                    src0->ne, src_cast_nb, GGML_MAX_DIMS);
+                aclnn_cast(ctx, acl_src0.get(), acl_src_cast.get(), ggml_cann_type_mapping(dst->type));
+
+                scatter_batched(src_cast_allocator.get(),
+                                ggml_cann_type_mapping(dst->type), ggml_type_size(dst->type),
+                                src_cast_nb);
                 break;
             }
         default:
@@ -3268,29 +3470,50 @@ void ggml_cann_pad_reflect_1d(ggml_backend_cann_context & ctx, ggml_tensor * dst
     int64_t           paddingsArray[2] = { opts[0], opts[1] };
     acl_int_array_ptr paddings         = ggml_cann_create_int_array(paddingsArray, 2);
 
-    for (int64_t i = 0; i < src0->ne[3]; i++) {
-        acl_tensor_ptr acl_src =
-            ggml_cann_create_tensor((char *) src0->data + i * src0->ne[3], ggml_cann_type_mapping(src0->type),
-                                    ggml_element_size(src0), src0->ne, src0->nb, 3);
+    // Collapsing ne[2]*ne[3] into a single batch dimension requires that dim3
+    // is contiguous with respect to dim2 in both src and dst.
+    GGML_ASSERT(src0->nb[3] == src0->nb[2] * src0->ne[2]);
+    GGML_ASSERT(dst->nb[3]  == dst->nb[2]  * dst->ne[2]);
 
-        acl_tensor_ptr acl_dst =
-            ggml_cann_create_tensor((char *) dst->data + i * src0->ne[3], ggml_cann_type_mapping(dst->type),
-                                    ggml_element_size(dst), dst->ne, dst->nb, 3);
+    int64_t src_ne_3d[3] = { src0->ne[0], src0->ne[1], src0->ne[2] * src0->ne[3] };
+    int64_t dst_ne_3d[3] = { dst->ne[0],  dst->ne[1],  dst->ne[2]  * dst->ne[3]  };
 
-        GGML_CANN_CALL_ACLNN_OP(ctx, ReflectionPad1d, acl_src.get(), paddings.get(), acl_dst.get());
-    }
+    acl_tensor_ptr acl_src = ggml_cann_create_tensor(src0->data, ggml_cann_type_mapping(src0->type),
+                                                     ggml_element_size(src0), src_ne_3d, src0->nb, 3);
+
+    acl_tensor_ptr acl_dst = ggml_cann_create_tensor(dst->data, ggml_cann_type_mapping(dst->type),
+                                                     ggml_element_size(dst), dst_ne_3d, dst->nb, 3);
+
+    GGML_CANN_CALL_ACLNN_OP(ctx, ReflectionPad1d, acl_src.get(), paddings.get(), acl_dst.get());
 }
 
 void ggml_cann_count_equal(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
     ggml_tensor * src0 = dst->src[0];
     ggml_tensor * src1 = dst->src[1];
 
+    // Write element-wise equality (0 or 1) into a temporary buffer to avoid
+    // modifying src0 in-place.  Use the same type as src0 so ReduceSum can
+    // consume it directly without a type cast.
+    ggml_cann_pool_alloc eq_alloc(ctx.pool(), ggml_nelements(src0) * ggml_element_size(src0));
+    size_t eq_nb[GGML_MAX_DIMS];
+    eq_nb[0] = ggml_element_size(src0);
+    for (int i = 1; i < GGML_MAX_DIMS; i++) {
+        eq_nb[i] = eq_nb[i - 1] * src0->ne[i - 1];
+    }
+    acl_tensor_ptr acl_eq = ggml_cann_create_tensor(
+        eq_alloc.get(), ggml_cann_type_mapping(src0->type), ggml_element_size(src0),
+        src0->ne, eq_nb, GGML_MAX_DIMS);
+
     acl_tensor_ptr acl_self  = ggml_cann_create_tensor(src0);
     acl_tensor_ptr acl_other = ggml_cann_create_tensor(src1);
+    GGML_CANN_CALL_ACLNN_OP(ctx, EqTensor, acl_self.get(), acl_other.get(), acl_eq.get());
 
-    GGML_CANN_CALL_ACLNN_OP(ctx, InplaceEqTensor, acl_self.get(), acl_other.get());
-
-    ggml_cann_sum(ctx, dst);
+    // Sum the 0/1 values into dst.
+    acl_tensor_ptr    acl_dst    = ggml_cann_create_tensor(dst);
+    int64_t           dims[4]    = { 0, 1, 2, 3 };
+    acl_int_array_ptr dims_arr   = ggml_cann_create_int_array(dims, 4);
+    GGML_CANN_CALL_ACLNN_OP(ctx, ReduceSum, acl_eq.get(), dims_arr.get(), true,
+                            ggml_cann_type_mapping(dst->type), acl_dst.get());
 }
 
 void ggml_cann_step(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
@@ -3306,6 +3529,27 @@ void ggml_cann_step(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
     GGML_CANN_CALL_ACLNN_OP(ctx, GtScalar, acl_src.get(), alpha.get(), acl_dst.get());
 }
 
+void ggml_cann_softplus(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
+    ggml_tensor * src0 = dst->src[0];
+
+    acl_tensor_ptr acl_src = ggml_cann_create_tensor(src0);
+    acl_tensor_ptr acl_dst = ggml_cann_create_tensor(dst);
+
+    float          beta_val      = 1.0f;
+    float          threshold_val = 20.0f;
+    acl_scalar_ptr beta          = ggml_cann_create_scalar(&beta_val,      ACL_FLOAT);
+    acl_scalar_ptr threshold     = ggml_cann_create_scalar(&threshold_val, ACL_FLOAT);
+
+    GGML_CANN_CALL_ACLNN_OP(ctx, Softplus, acl_src.get(), beta.get(), threshold.get(), acl_dst.get());
+}
+
+void ggml_cann_geglu_quick(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
+    auto gelu_quick_fn = [](ggml_backend_cann_context & ctx, aclTensor * acl_src, aclTensor * acl_dst) {
+        GGML_CANN_CALL_ACLNN_OP(ctx, GeluV2, acl_src, 0, acl_dst);
+    };
+    ggml_cann_op_unary_gated(gelu_quick_fn, ctx, dst);
+}
+
 /**
  * @brief Performs expert-specific matrix multiplication (MoE) with
  * floating-point precision using the CANN backend.
@@ -3892,46 +4136,65 @@ void ggml_cann_flash_attn_ext(ggml_backend_cann_context & ctx, ggml_tensor * dst
 }
 
 static void ggml_cann_out_prod_fp(ggml_backend_cann_context & ctx, ggml_tensor * dst) {
-    ggml_tensor * src0 = dst->src[0];  // weight
-    ggml_tensor * src1 = dst->src[1];  // input
+    ggml_tensor * src0 = dst->src[0];  // weight  [ne00=m, ne01=K, ne02, ne03]
+    ggml_tensor * src1 = dst->src[1];  // input   [ne10=n, ne11=K, ne12, ne13]
     GGML_TENSOR_BINARY_OP_LOCALS
 
-    acl_tensor_ptr acl_dst = ggml_cann_create_tensor(dst);
-    GGML_CANN_CALL_ACLNN_OP(ctx, InplaceZero, acl_dst.get());
+    // dst[i,j] = sum_k src0[i,k] * src1[j,k]  i.e. dst = src0 @ src1^T.
+    //
+    // ggml_cann_create_tensor reverses dimension order, so ACL sees:
+    //   acl_src0 slice:   ggml[m,K]  ->  ACL[K,m]
+    //   acl_src1 slice:   ggml[n,K]  ->  ACL[K,n]
+    //   acl_dst  slice:   ggml[m,n]  ->  ACL[n,m]
+    //
+    // Build a transposed view of src1 by swapping ne[0]/ne[1]:
+    //   src1_t:  ggml[K,n] (swapped strides)  ->  ACL[n,K]
+    //
+    // Matmul(src1_t [n,K], src0 [K,m]) = [n,m] = acl_dst  ✓
+    //
+    // The outer batch loop is kept because src0 may have fewer batch slices than
+    // dst (ne02 <= ne2, ne03 <= ne3): this is a strided-broadcast not supported
+    // by standard CANN Matmul broadcasting.
+
+    const aclDataType src0_acl_type = ggml_cann_type_mapping(src0->type);
+    const aclDataType src1_acl_type = ggml_cann_type_mapping(src1->type);
+    const aclDataType dst_acl_type  = ggml_cann_type_mapping(dst->type);
+    const size_t      src0_type_sz  = ggml_type_size(src0->type);
+    const size_t      src1_type_sz  = ggml_type_size(src1->type);
+    const size_t      dst_type_sz   = ggml_type_size(dst->type);
 
     const int64_t dps2 = ne2 / ne02;
     const int64_t dps3 = ne3 / ne03;
+
     for (int64_t i3 = 0; i3 < ne3; i3++) {
         for (int64_t i2 = 0; i2 < ne2; i2++) {
             const int64_t i02 = i2 / dps2;
             const int64_t i03 = i3 / dps3;
 
-            const int64_t  i12 = i2;
-            const int64_t  i13 = i3;
-            acl_tensor_ptr accumulator =
-                ggml_cann_create_tensor((char *) dst->data + i2 * nb2 + i3 * nb3, ggml_cann_type_mapping(dst->type),
-                                        ggml_type_size(dst->type), dst->ne, dst->nb, 2);
+            // src0 2D slice at [i02, i03]: ggml [m, K] -> ACL [K, m]
+            int64_t src0_ne[2] = { ne00, ne01 };
+            size_t  src0_nb[2] = { nb00, nb01 };
+            acl_tensor_ptr acl_src0_s = ggml_cann_create_tensor(
+                (char *) src0->data + i02 * nb02 + i03 * nb03,
+                src0_acl_type, src0_type_sz, src0_ne, src0_nb, 2);
 
-            // The outer product needs to be accumulated in this dimension.
-            for (int64_t i1 = 0; i1 < ne11; i1++) {
-                acl_tensor_ptr acl_input = ggml_cann_create_tensor(
-                    (char *) src1->data + i1 * nb11 + i12 * nb12 + i13 * nb13, ggml_cann_type_mapping(src0->type),
-                    ggml_type_size(src0->type), src1->ne, src1->nb, 1);
+            // src1 transposed 2D slice at [i2, i3]: swap ne/nb -> ggml[K,n] -> ACL[n,K]
+            int64_t src1_t_ne[2] = { ne11, ne10 };
+            size_t  src1_t_nb[2] = { nb11, nb10 };
+            acl_tensor_ptr acl_src1_t = ggml_cann_create_tensor(
+                (char *) src1->data + i2 * nb12 + i3 * nb13,
+                src1_acl_type, src1_type_sz, src1_t_ne, src1_t_nb, 2);
 
-                acl_tensor_ptr acl_weight = ggml_cann_create_tensor(
-                    (char *) src0->data + i1 * nb01 + i02 * nb02 + i03 * nb03, ggml_cann_type_mapping(src0->type),
-                    ggml_type_size(src0->type), src0->ne, src0->nb, 1);
+            // dst 2D slice at [i2, i3]: ggml [m, n] -> ACL [n, m]
+            int64_t dst_ne[2] = { ne0, ne1 };
+            size_t  dst_nb[2] = { nb0, nb1 };
+            acl_tensor_ptr acl_dst_s = ggml_cann_create_tensor(
+                (char *) dst->data + i2 * nb2 + i3 * nb3,
+                dst_acl_type, dst_type_sz, dst_ne, dst_nb, 2);
 
-                ggml_cann_pool_alloc output_allocator(ctx.pool());
-                void *               output_buffer = output_allocator.alloc(ggml_nbytes(dst));
-                acl_tensor_ptr       acl_out = ggml_cann_create_tensor(output_buffer, ggml_cann_type_mapping(dst->type),
-                                                                       ggml_type_size(dst->type), dst->ne, dst->nb, 2);
-
-                GGML_CANN_CALL_ACLNN_OP(ctx, Ger, acl_input.get(), acl_weight.get(), acl_out.get());
-                float       alpha_value = 1.0f;
-                aclScalar * alpha       = aclCreateScalar(&alpha_value, ACL_FLOAT);
-                GGML_CANN_CALL_ACLNN_OP(ctx, InplaceAdd, accumulator.get(), acl_out.get(), alpha);
-            }
+            // Matmul(src1_t [n,K], src0 [K,m]) = [n,m] = acl_dst_s  ✓
+            GGML_CANN_CALL_ACLNN_OP(ctx, Matmul,
+                acl_src1_t.get(), acl_src0_s.get(), acl_dst_s.get(), (int8_t) 1);
         }
     }
 }
@@ -4170,3 +4433,4 @@ void ggml_cann_gated_linear_attn(ggml_backend_cann_context & ctx, ggml_tensor *
         }
     }
 }
+

ggml/src/ggml-cann/aclnn_ops.h

@@ -32,6 +32,9 @@
 #include <aclnnop/aclnn_cat.h>
 #include <aclnnop/aclnn_clamp.h>
 #include <aclnnop/aclnn_cos.h>
+#include <aclnnop/aclnn_cumsum.h>
+#include <aclnnop/aclnn_tril.h>
+#include <aclnnop/aclnn_triu.h>
 #include <aclnnop/aclnn_exp.h>
 #include <aclnnop/aclnn_gelu.h>
 #include <aclnnop/aclnn_gelu_v2.h>
@@ -47,6 +50,9 @@
 #include <aclnnop/aclnn_sign.h>
 #include <aclnnop/aclnn_silu.h>
 #include <aclnnop/aclnn_sin.h>
+#include <aclnnop/aclnn_softplus.h>
+#include <aclnnop/aclnn_swi_glu.h>
+#include <aclnnop/aclnn_geglu.h>
 #include <aclnnop/aclnn_slice.h>
 #include <aclnnop/aclnn_sqrt.h>
 #include <aclnnop/aclnn_tanh.h>
@@ -69,6 +75,9 @@
  */
 void ggml_cann_repeat(ggml_backend_cann_context & ctx, ggml_tensor * dst);
 
+void ggml_cann_swiglu(ggml_backend_cann_context & ctx, ggml_tensor * dst);
+void ggml_cann_geglu(ggml_backend_cann_context & ctx, ggml_tensor * dst, int64_t approximate);
+
 /**
  * @brief   Applies the Leaky ReLU activation function to a tensor using the CANN
  *          backend.
@@ -325,6 +334,48 @@ void ggml_cann_sum_rows(ggml_backend_cann_context & ctx, ggml_tensor * dst);
 
 void ggml_cann_sum(ggml_backend_cann_context & ctx, ggml_tensor * dst);
 
+/**
+ * @brief   Computes the cumulative sum of a ggml tensor along dim 0 using the
+ *          CANN backend.
+ *
+ * @param ctx The CANN context used for operations.
+ * @param dst The destination tensor. dst->op is `GGML_OP_CUMSUM`.
+ */
+void ggml_cann_cumsum(ggml_backend_cann_context & ctx, ggml_tensor * dst);
+
+/**
+ * @brief   Computes a triangular mask (tril/triu) of a square ggml tensor
+ *          using the CANN backend.
+ *
+ * @param ctx The CANN context used for operations.
+ * @param dst The destination tensor. dst->op is `GGML_OP_TRI`.
+ */
+void ggml_cann_tri(ggml_backend_cann_context & ctx, ggml_tensor * dst);
+
+/**
+ * @brief   Solves a triangular linear system AX=B using the CANN backend.
+ *
+ * @param ctx The CANN context used for operations.
+ * @param dst The destination tensor. dst->op is `GGML_OP_SOLVE_TRI`.
+ */
+void ggml_cann_solve_tri(ggml_backend_cann_context & ctx, ggml_tensor * dst);
+
+/**
+ * @brief   Creates a diagonal matrix from a vector using the CANN backend.
+ *
+ * @param ctx The CANN context used for operations.
+ * @param dst The destination tensor. dst->op is `GGML_OP_DIAG`.
+ */
+void ggml_cann_diag(ggml_backend_cann_context & ctx, ggml_tensor * dst);
+
+/**
+ * @brief   Fills a tensor with a constant scalar value using the CANN backend.
+ *
+ * @param ctx The CANN context used for operations.
+ * @param dst The destination tensor. dst->op is `GGML_OP_FILL`.
+ */
+void ggml_cann_fill(ggml_backend_cann_context & ctx, ggml_tensor * dst);
+
 /**
  * @brief   Upsamples a ggml tensor using nearest neighbor interpolation using
  *          the CANN backend.
@@ -461,6 +512,9 @@ void ggml_cann_timestep_embedding(ggml_backend_cann_context & ctx, ggml_tensor *
 // @see ggml_cann_dup.
 void ggml_cann_cpy(ggml_backend_cann_context & ctx, ggml_tensor * dst);
 
+// @see ggml_cann_acc, but copies src1 into dst instead of adding.
+void ggml_cann_set(ggml_backend_cann_context & ctx, ggml_tensor * dst);
+
 /**
  * @brief   Computes the softmax activation with optional masking.
  *
@@ -813,6 +867,8 @@ void ggml_cann_count_equal(ggml_backend_cann_context & ctx, ggml_tensor * dst);
  *            dst->op is expected to be `GGML_OP_STEP`.
  */
 void ggml_cann_step(ggml_backend_cann_context & ctx, ggml_tensor * dst);
+void ggml_cann_softplus(ggml_backend_cann_context & ctx, ggml_tensor * dst);
+void ggml_cann_geglu_quick(ggml_backend_cann_context & ctx, ggml_tensor * dst);
 
 /**
  * @brief   Performs the Flash Attention extended operator using the CANN backend.

ggml/src/ggml-cann/ggml-cann.cpp

@@ -1428,6 +1428,22 @@ static bool ggml_backend_cann_buffer_cpy_tensor(ggml_backend_buffer_t buffer,
     return false;
 }
 
+/**
+ * @brief Set a region of a tensor's device memory to a specified value.
+ *
+ * @param buffer The CANN buffer containing the tensor.
+ * @param tensor Pointer to the tensor whose memory will be set.
+ * @param value The value to which each byte in the region will be set.
+ * @param offset Byte offset within the tensor's data to start setting.
+ * @param size Number of bytes to set.
+ */
+static void ggml_backend_cann_buffer_memset_tensor(ggml_backend_buffer_t buffer, ggml_tensor * tensor, uint8_t value, size_t offset, size_t size) {
+    ggml_backend_cann_buffer_context * ctx = (ggml_backend_cann_buffer_context *) buffer->context;
+
+    ggml_cann_set_device(ctx->device);
+    ACL_CHECK(aclrtMemset((char *) tensor->data + offset, size, value, size));
+}
+
 /**
  * @brief Clear a CANN buffer by setting all its memory to a specified value.
  *
@@ -1454,7 +1470,7 @@ static const ggml_backend_buffer_i ggml_backend_cann_buffer_interface = {
     /* .free_buffer     = */ ggml_backend_cann_buffer_free_buffer,
     /* .get_base        = */ ggml_backend_cann_buffer_get_base,
     /* .init_tensor     = */ ggml_backend_cann_buffer_init_tensor,
-    /* .memset_tensor   = */ NULL,
+    /* .memset_tensor   = */ ggml_backend_cann_buffer_memset_tensor,
     /* .set_tensor      = */ ggml_backend_cann_buffer_set_tensor,
     /* .get_tensor      = */ ggml_backend_cann_buffer_get_tensor,
     /* .set_tensor_2d   = */ NULL,
@@ -1835,6 +1851,9 @@ static bool ggml_cann_compute_forward(ggml_backend_cann_context & ctx, struct gg
                 case GGML_UNARY_OP_STEP:
                     ggml_cann_step(ctx, dst);
                     break;
+                case GGML_UNARY_OP_SOFTPLUS:
+                    ggml_cann_softplus(ctx, dst);
+                    break;
                 default:
                     return false;
             }
@@ -1845,20 +1864,16 @@ static bool ggml_cann_compute_forward(ggml_backend_cann_context & ctx, struct gg
                     GGML_CANN_CALL_OP_UNARY_GATED(Relu);
                     break;
                 case GGML_GLU_OP_GEGLU:
+                    ggml_cann_geglu(ctx, dst, 0);  // approximate=0 → tanh
+                    break;
                 case GGML_GLU_OP_GEGLU_ERF:
-                    // aclnnGelu internally uses the erf-based approximation.
-                    GGML_CANN_CALL_OP_UNARY_GATED(Gelu);
+                    ggml_cann_geglu(ctx, dst, 1);  // approximate=1 → erf
                     break;
                 case GGML_GLU_OP_SWIGLU:
-                    GGML_CANN_CALL_OP_UNARY_GATED(Silu);
+                    ggml_cann_swiglu(ctx, dst);
                     break;
                 case GGML_GLU_OP_GEGLU_QUICK:
-                    {
-                        auto lambda = [](ggml_backend_cann_context & ctx, aclTensor * acl_src, aclTensor * acl_dst) {
-                            GGML_CANN_CALL_ACLNN_OP(ctx, GeluV2, acl_src, 0, acl_dst);
-                        };
-                        ggml_cann_op_unary_gated(lambda, ctx, dst);
-                    }
+                    ggml_cann_geglu_quick(ctx, dst);
                     break;
                 default:
                     return false;
@@ -1920,6 +1935,9 @@ static bool ggml_cann_compute_forward(ggml_backend_cann_context & ctx, struct gg
         case GGML_OP_CPY:
             ggml_cann_cpy(ctx, dst);
             break;
+        case GGML_OP_SET:
+            ggml_cann_set(ctx, dst);
+            break;
         case GGML_OP_CONT:
             ggml_cann_dup(ctx, dst);
             break;
@@ -1989,6 +2007,21 @@ static bool ggml_cann_compute_forward(ggml_backend_cann_context & ctx, struct gg
         case GGML_OP_SSM_CONV:
             ggml_cann_ssm_conv(ctx, dst);
             break;
+        case GGML_OP_CUMSUM:
+            ggml_cann_cumsum(ctx, dst);
+            break;
+        case GGML_OP_TRI:
+            ggml_cann_tri(ctx, dst);
+            break;
+        case GGML_OP_FILL:
+            ggml_cann_fill(ctx, dst);
+            break;
+        case GGML_OP_DIAG:
+            ggml_cann_diag(ctx, dst);
+            break;
+        case GGML_OP_SOLVE_TRI:
+            ggml_cann_solve_tri(ctx, dst);
+            break;
         default:
             return false;
     }
@@ -2324,6 +2357,7 @@ static enum ggml_status ggml_backend_cann_graph_compute(ggml_backend_t backend,
     if (use_cann_graph) {
         // If no matching graph is found, the graph needs to be recaptured.
         graph_capture_required = !cann_ctx->graph_lru_cache.find_and_move_to_front(cgraph);
+
         if (graph_capture_required) {
             // If no matching graph is found, add a new ACL graph.
             ggml_cann_graph * new_graph = ggml_cann_graph::create_from_cgraph(cgraph);
@@ -2382,6 +2416,7 @@ static bool ggml_backend_cann_supports_op(ggml_backend_dev_t dev, const ggml_ten
                 case GGML_UNARY_OP_SGN:
                 case GGML_UNARY_OP_STEP:
                 case GGML_UNARY_OP_GELU_ERF:
+                case GGML_UNARY_OP_SOFTPLUS:
                     return true;
                 default:
                     return false;
@@ -2572,6 +2607,7 @@ static bool ggml_backend_cann_supports_op(ggml_backend_dev_t dev, const ggml_ten
         case GGML_OP_SUM_ROWS:
         case GGML_OP_ARGSORT:
         case GGML_OP_ACC:
+        case GGML_OP_SET:
         case GGML_OP_GROUP_NORM:
             return true;
         case GGML_OP_PAD:
@@ -2649,6 +2685,16 @@ static bool ggml_backend_cann_supports_op(ggml_backend_dev_t dev, const ggml_ten
             }
         case GGML_OP_SSM_CONV:
             return true;
+        case GGML_OP_CUMSUM:
+            return op->src[0]->type == GGML_TYPE_F32;
+        case GGML_OP_TRI:
+            return op->src[0]->type == GGML_TYPE_F32;
+        case GGML_OP_FILL:
+            return op->src[0]->type == GGML_TYPE_F32;
+        case GGML_OP_DIAG:
+            return op->src[0]->type == GGML_TYPE_F32;
+        case GGML_OP_SOLVE_TRI:
+            return op->src[0]->type == GGML_TYPE_F32;
         default:
             return false;
     }
@@ -2700,8 +2746,8 @@ static const ggml_backend_i ggml_backend_cann_interface = {
     /* .free                    = */ ggml_backend_cann_free,
     /* .set_tensor_async        = */ ggml_backend_cann_set_tensor_async,
     /* .get_tensor_async        = */ ggml_backend_cann_get_tensor_async,
-    /* .get_tensor_2d_async     = */ NULL,
     /* .set_tensor_2d_async     = */ NULL,
+    /* .get_tensor_2d_async     = */ NULL,
     /* .cpy_tensor_async        = */ ggml_backend_cann_cpy_tensor_async,
     /* .synchronize             = */ ggml_backend_cann_synchronize,
     /* .graph_plan_create       = */ NULL,

ggml/src/ggml-cpu/CMakeLists.txt

@@ -485,6 +485,13 @@ function(ggml_add_cpu_backend_variant_impl tag_name)
             if (GGML_RV_ZIHINTPAUSE)
                 string(APPEND MARCH_STR "_zihintpause")
             endif()
+            if (GGML_CPU_RISCV64_SPACEMIT)
+                # `xsmtvdotii' is only required for GCC >= 15.
+                if (CMAKE_C_COMPILER_ID STREQUAL "GNU" AND
+                    CMAKE_C_COMPILER_VERSION VERSION_GREATER_EQUAL 15)
+                    string(APPEND MARCH_STR "_xsmtvdotii")
+                endif()
+            endif()
 
             list(APPEND ARCH_FLAGS "-march=${MARCH_STR}" -mabi=lp64d)
         else()

ggml/src/ggml-cpu/amx/mmq.cpp

@@ -2005,12 +2005,12 @@ void tinygemm_kernel_amx(int M, int N, int KB, const void * RESTRICT _A, const v
     const int lda = KB * sizeof(TA);
     //const int ldb = KB * sizeof(TB);
 
-    static thread_local packed_B_t Tile0[TILE_N * TILE_K];
-    static thread_local packed_B_t Tile1[TILE_N * TILE_K];
-    static thread_local int8_t Tile23[TILE_M * TILE_K];
+    alignas(64) static thread_local packed_B_t Tile0[TILE_N * TILE_K];
+    alignas(64) static thread_local packed_B_t Tile1[TILE_N * TILE_K];
+    alignas(64) static thread_local int8_t Tile23[TILE_M * TILE_K];
 
-    static thread_local int32_t TileC0[TILE_M * TILE_N * 4];
-    static thread_local int32_t TileC1[TILE_M * TILE_N * 4];
+    alignas(64) static thread_local int32_t TileC0[TILE_M * TILE_N * 4];
+    alignas(64) static thread_local int32_t TileC1[TILE_M * TILE_N * 4];
 
     // double buffering C to interleave avx512 and amx
     int32_t * C_cur = TileC0;
@@ -2187,21 +2187,21 @@ void tinygemm_kernel_amx(int M, int N, int KB, const void * RESTRICT _A, const v
     const int m1 = std::max(M - TILE_M, 0);
     //const int lda = KB * sizeof(TA);
 
-    static thread_local int8_t Tile0[TILE_N * TILE_K];
-    static thread_local int8_t Tile1[TILE_N * TILE_K];
-    static thread_local int8_t Tile23[TILE_M * TILE_K];
+    alignas(64) static thread_local int8_t Tile0[TILE_N * TILE_K];
+    alignas(64) static thread_local int8_t Tile1[TILE_N * TILE_K];
+    alignas(64) static thread_local int8_t Tile23[TILE_M * TILE_K];
 
     // mat mul result for each group
-    static thread_local int32_t Tile4[TILE_M * TILE_N];
-    static thread_local int32_t Tile5[TILE_M * TILE_N];
-    static thread_local int32_t Tile6[TILE_M * TILE_N];
-    static thread_local int32_t Tile7[TILE_M * TILE_N];
+    alignas(64) static thread_local int32_t Tile4[TILE_M * TILE_N];
+    alignas(64) static thread_local int32_t Tile5[TILE_M * TILE_N];
+    alignas(64) static thread_local int32_t Tile6[TILE_M * TILE_N];
+    alignas(64) static thread_local int32_t Tile7[TILE_M * TILE_N];
 
     // sum of each QK_K block, contains 8 groups, int32
-    static thread_local int32_t Sumi4[TILE_M * TILE_N];
-    static thread_local int32_t Sumi5[TILE_M * TILE_N];
-    static thread_local int32_t Sumi6[TILE_M * TILE_N];
-    static thread_local int32_t Sumi7[TILE_M * TILE_N];
+    alignas(64) static thread_local int32_t Sumi4[TILE_M * TILE_N];
+    alignas(64) static thread_local int32_t Sumi5[TILE_M * TILE_N];
+    alignas(64) static thread_local int32_t Sumi6[TILE_M * TILE_N];
+    alignas(64) static thread_local int32_t Sumi7[TILE_M * TILE_N];
 
     const int k_group_size = std::is_same<TB, block_q6_K>::value ? 16 : 32;
     for (int i = 0; i < KB; ++i) {

ggml/src/ggml-cpu/arch/arm/repack.cpp

@@ -5023,6 +5023,71 @@ void ggml_gemm_q8_0_4x8_q8_0(int                        n,
     UNUSED(ncols_interleaved);
     UNUSED(blocklen);
 
+#if defined(__aarch64__) && defined(__ARM_FEATURE_SVE) && defined(__ARM_FEATURE_MATMUL_INT8)
+    if (svcntb() * 8 == 256) {
+        const block_q8_0x4 * b_ptr_base = (const block_q8_0x4 *) vx;
+
+        static const uint32_t idx_arr[8] = {0, 1, 4, 5, 2, 3, 6, 7};
+        svuint32_t idx = svld1(svptrue_b32(), idx_arr);
+        static const uint32_t idx_arr1[8] = {0, 1, 2, 3, 1, 2, 3, 0};
+        svuint32_t idx_sc1 = svld1(svptrue_b32(), idx_arr1);
+        static const uint32_t idx_arr2[8] = {0, 1, 2, 3, 0, 1, 2, 3};
+        svuint32_t idx_sc2 = svld1(svptrue_b32(), idx_arr2);
+
+        for (int y = 0; y < nr; y += 4) {
+            const block_q8_0x4 * a_ptr_base = (const block_q8_0x4 *) vy + (y / 4) * nb;
+
+            for (int x = 0; x < nc; x += ncols_interleaved) {
+                const block_q8_0x4 * b_ptr = b_ptr_base + (x / 4) * nb;
+                const block_q8_0x4 * a_ptr = a_ptr_base;
+
+                svfloat32_t acc_f32_01 = svdup_f32(0);
+                svfloat32_t acc_f32_23 = svdup_f32(0);
+
+                for (int b = 0; b < nb; b++) {
+
+                    svint32_t acc_01 = svdup_s32(0);
+                    svint32_t acc_23 = svdup_s32(0);
+
+                    // Process 4 chunks of 8 positions each
+                    for (int chunk = 0; chunk < 4; chunk++) {
+                        svint8_t s_a01 = svld1rq_s8(svptrue_b8(), a_ptr->qs + chunk * 32);
+                        svint8_t s_a23 = svld1rq_s8(svptrue_b8(), a_ptr->qs + chunk * 32 + 16);
+                        svint8_t s_b0123 = svld1_s8(svptrue_b8(), b_ptr->qs + chunk * 32);
+
+                        acc_01 = svmmla_s32(acc_01, s_a01, s_b0123);
+                        acc_23 = svmmla_s32(acc_23, s_a23, s_b0123);
+                    }
+
+                    // Reorder outputs from 2×2 tiles to row-major
+                    // acc[01] = [r0c0, r0c1, r1c0, r1c1, r0c2, r0c3, r1c2, r1c3]
+                    // acc[23] = [r2c0, r2c1, r3c0, r3c1, r2c2, r2c3, r3c2, r3c3]
+
+                    svint32_t row01 = svtbl_s32(acc_01, idx);
+                    svint32_t row23 = svtbl_s32(acc_23, idx);
+
+                    svfloat16_t temp1 = svld1_f16(svptrue_pat_b16(SV_VL4), (const __fp16 *) a_ptr->d);
+                    svfloat16_t temp2 = svld1_f16(svptrue_pat_b16(SV_VL4), (const __fp16 *) b_ptr->d);
+                    svfloat32_t sv_a_d = svtbl_f32(svcvt_f32_f16_x(svptrue_b32(), svzip1_f16(temp1, temp1)), idx_sc1);
+                    svfloat32_t sv_b_d = svtbl_f32(svcvt_f32_f16_x(svptrue_b32(), svzip1_f16(temp2, temp2)), idx_sc2);
+
+                    acc_f32_01 = svmla_f32_x(svptrue_b32(), acc_f32_01, svcvt_f32_s32_x(svptrue_b32(), row01), svmul_lane_f32(sv_b_d, sv_a_d, 0));
+                    acc_f32_23 = svmla_f32_x(svptrue_b32(), acc_f32_23, svcvt_f32_s32_x(svptrue_b32(), row23), svmul_lane_f32(sv_b_d, sv_a_d, 2));
+                    a_ptr++;
+                    b_ptr++;
+                }
+
+                svbool_t pg4 = svptrue_pat_b32(SV_VL4);
+                svst1_f32(pg4, s + (y+0) * bs + x, acc_f32_01);
+                svst1_f32(pg4, s + (y+1) * bs + x, svext_f32(acc_f32_01, acc_f32_01, 4));
+                svst1_f32(pg4, s + (y+2) * bs + x, acc_f32_23);
+                svst1_f32(pg4, s + (y+3) * bs + x, svext_f32(acc_f32_23, acc_f32_23, 4));
+            }
+        }
+        return;
+    }
+#endif  // SVE compile-time end
+
 #if defined(__aarch64__) && defined(__ARM_NEON) && defined(__ARM_FEATURE_MATMUL_INT8)
     const block_q8_0x4 * b_ptr_base = (const block_q8_0x4 *) vx;
 

ggml/src/ggml-cpu/arch/x86/quants.c

@@ -2300,9 +2300,8 @@ void ggml_vec_dot_q6_K_q8_K(int n, float * GGML_RESTRICT s, size_t bs, const voi
 
 #if defined __AVX2__
 
-    const __m256i m4 = _mm256_set1_epi8(0xF);
-    const __m256i m2 = _mm256_set1_epi8(3);
-    const __m256i m32s = _mm256_set1_epi8(32);
+    const __m256i m3 = _mm256_set1_epi8(3);
+    const __m256i m15 = _mm256_set1_epi8(15);
 
     __m256 acc = _mm256_setzero_ps();
 
@@ -2314,53 +2313,45 @@ void ggml_vec_dot_q6_K_q8_K(int n, float * GGML_RESTRICT s, size_t bs, const voi
         const uint8_t * GGML_RESTRICT qh = x[i].qh;
         const int8_t  * GGML_RESTRICT q8 = y[i].qs;
 
+        const __m256i q8sums = _mm256_loadu_si256((const __m256i*)y[i].bsums);
         const __m128i scales = _mm_loadu_si128((const __m128i*)x[i].scales);
+        const __m256i scales_16 = _mm256_cvtepi8_epi16(scales);
+        const __m256i q8sclsub = _mm256_slli_epi32(_mm256_madd_epi16(q8sums, scales_16), 5);
 
         __m256i sumi = _mm256_setzero_si256();
 
         int is = 0;
 
         for (int j = 0; j < QK_K/128; ++j) {
-
-            const __m128i scale_0 = _mm_shuffle_epi8(scales, get_scale_shuffle(is + 0));
-            const __m128i scale_1 = _mm_shuffle_epi8(scales, get_scale_shuffle(is + 1));
-            const __m128i scale_2 = _mm_shuffle_epi8(scales, get_scale_shuffle(is + 2));
-            const __m128i scale_3 = _mm_shuffle_epi8(scales, get_scale_shuffle(is + 3));
-            is += 4;
-
             const __m256i q4bits1 = _mm256_loadu_si256((const __m256i*)q4); q4 += 32;
             const __m256i q4bits2 = _mm256_loadu_si256((const __m256i*)q4); q4 += 32;
             const __m256i q4bitsH = _mm256_loadu_si256((const __m256i*)qh); qh += 32;
 
-            const __m256i q4h_0 = _mm256_slli_epi16(_mm256_and_si256(q4bitsH, m2), 4);
-            const __m256i q4h_1 = _mm256_slli_epi16(_mm256_and_si256(_mm256_srli_epi16(q4bitsH, 2), m2), 4);
-            const __m256i q4h_2 = _mm256_slli_epi16(_mm256_and_si256(_mm256_srli_epi16(q4bitsH, 4), m2), 4);
-            const __m256i q4h_3 = _mm256_slli_epi16(_mm256_and_si256(_mm256_srli_epi16(q4bitsH, 6), m2), 4);
+            const __m256i q4h_0 = _mm256_slli_epi16(_mm256_and_si256(q4bitsH, m3), 4);
+            const __m256i q4h_1 = _mm256_slli_epi16(_mm256_and_si256(q4bitsH, _mm256_set1_epi8(12)), 2);
+            const __m256i q4h_2 = _mm256_and_si256(q4bitsH, _mm256_set1_epi8(48));
+            const __m256i q4h_3 = _mm256_srli_epi16(_mm256_and_si256(q4bitsH, _mm256_set1_epi8(-64)), 2);
 
-            const __m256i q4_0 = _mm256_or_si256(_mm256_and_si256(q4bits1, m4), q4h_0);
-            const __m256i q4_1 = _mm256_or_si256(_mm256_and_si256(q4bits2, m4), q4h_1);
-            const __m256i q4_2 = _mm256_or_si256(_mm256_and_si256(_mm256_srli_epi16(q4bits1, 4), m4), q4h_2);
-            const __m256i q4_3 = _mm256_or_si256(_mm256_and_si256(_mm256_srli_epi16(q4bits2, 4), m4), q4h_3);
+            const __m256i q4_0 = _mm256_or_si256(_mm256_and_si256(q4bits1, m15), q4h_0);
+            const __m256i q4_1 = _mm256_or_si256(_mm256_and_si256(q4bits2, m15), q4h_1);
+            const __m256i q4_2 = _mm256_or_si256(_mm256_and_si256(_mm256_srli_epi16(q4bits1, 4), m15), q4h_2);
+            const __m256i q4_3 = _mm256_or_si256(_mm256_and_si256(_mm256_srli_epi16(q4bits2, 4), m15), q4h_3);
 
             const __m256i q8_0 = _mm256_loadu_si256((const __m256i*)q8); q8 += 32;
             const __m256i q8_1 = _mm256_loadu_si256((const __m256i*)q8); q8 += 32;
             const __m256i q8_2 = _mm256_loadu_si256((const __m256i*)q8); q8 += 32;
             const __m256i q8_3 = _mm256_loadu_si256((const __m256i*)q8); q8 += 32;
 
-            __m256i q8s_0 = _mm256_maddubs_epi16(m32s, q8_0);
-            __m256i q8s_1 = _mm256_maddubs_epi16(m32s, q8_1);
-            __m256i q8s_2 = _mm256_maddubs_epi16(m32s, q8_2);
-            __m256i q8s_3 = _mm256_maddubs_epi16(m32s, q8_3);
-
             __m256i p16_0 = _mm256_maddubs_epi16(q4_0, q8_0);
             __m256i p16_1 = _mm256_maddubs_epi16(q4_1, q8_1);
             __m256i p16_2 = _mm256_maddubs_epi16(q4_2, q8_2);
             __m256i p16_3 = _mm256_maddubs_epi16(q4_3, q8_3);
 
-            p16_0 = _mm256_sub_epi16(p16_0, q8s_0);
-            p16_1 = _mm256_sub_epi16(p16_1, q8s_1);
-            p16_2 = _mm256_sub_epi16(p16_2, q8s_2);
-            p16_3 = _mm256_sub_epi16(p16_3, q8s_3);
+            const __m128i scale_0 = _mm_shuffle_epi8(scales, get_scale_shuffle(is + 0));
+            const __m128i scale_1 = _mm_shuffle_epi8(scales, get_scale_shuffle(is + 1));
+            const __m128i scale_2 = _mm_shuffle_epi8(scales, get_scale_shuffle(is + 2));
+            const __m128i scale_3 = _mm_shuffle_epi8(scales, get_scale_shuffle(is + 3));
+            is += 4;
 
             p16_0 = _mm256_madd_epi16(_mm256_cvtepi8_epi16(scale_0), p16_0);
             p16_1 = _mm256_madd_epi16(_mm256_cvtepi8_epi16(scale_1), p16_1);
@@ -2372,6 +2363,7 @@ void ggml_vec_dot_q6_K_q8_K(int n, float * GGML_RESTRICT s, size_t bs, const voi
 
         }
 
+        sumi = _mm256_sub_epi32(sumi, q8sclsub);
         acc = _mm256_fmadd_ps(_mm256_broadcast_ss(&d), _mm256_cvtepi32_ps(sumi), acc);
     }
 

ggml/src/ggml-cpu/ggml-cpu.cpp

@@ -195,8 +195,8 @@ static const struct ggml_backend_i ggml_backend_cpu_i = {
     /* .free                    = */ ggml_backend_cpu_free,
     /* .set_tensor_async        = */ NULL,
     /* .get_tensor_async        = */ NULL,
-    /* .get_tensor_2d_async     = */ NULL,
     /* .set_tensor_2d_async     = */ NULL,
+    /* .get_tensor_2d_async     = */ NULL,
     /* .cpy_tensor_async        = */ NULL,
     /* .synchronize             = */ NULL,
     /* .graph_plan_create       = */ ggml_backend_cpu_graph_plan_create,

ggml/src/ggml-cpu/llamafile/sgemm.cpp

@@ -2321,6 +2321,9 @@ class tinyBLAS_Q0_PPC {
     }
 
     void matmul(int64_t m, int64_t n) {
+    #if defined(_AIX) || defined(__BIG_ENDIAN__)
+        mnpack(0, m, 0, n);
+    #else
         const int64_t mc = 64;
         const int64_t kc = 64;
         int64_t nc = 64;
@@ -2334,7 +2337,6 @@ class tinyBLAS_Q0_PPC {
         } else {
             n_aligned = (n / 64) * 64;
         }
-
         if (n_aligned > 0) {
             if (n_aligned % 64 == 0)      nc = 64;
             else if (n_aligned == n)      nc = n;
@@ -2352,6 +2354,7 @@ class tinyBLAS_Q0_PPC {
         } else {
             mnpack(0, m, 0, n);
         }
+    #endif
     }
 
   private:
@@ -3191,12 +3194,16 @@ class tinyBLAS_PPC {
     }
 
     void matmul(int64_t m, int64_t n) {
+    #if defined(_AIX) || defined(__BIG_ENDIAN__)
+        mnpack(0, m, 0, n);
+    #else
         int64_t mc = 256; int64_t nc = 256; int64_t kc = 256;
         if (m % mc == 0 && n % nc == 0 && k % kc == 0) {
             matmul_tiled(m, n, mc, nc, kc);
         } else {
             mnpack(0, m, 0, n);
         }
+    #endif
     }
 
   private:

ggml/src/ggml-cpu/vec.h

@@ -1036,12 +1036,12 @@ inline static float ggml_gelu_quick_f32(float x) {
     return x*(1.0f/(1.0f+expf(GELU_QUICK_COEF*x)));
 }
 
-//inline static void ggml_vec_gelu_quick_f16(const int n, ggml_fp16_t * y, const ggml_fp16_t * x) {
-//    const uint16_t * i16 = (const uint16_t *) x;
-//    for (int i = 0; i < n; ++i) {
-//        y[i] = ggml_table_gelu_quick_f16[i16[i]];
-//    }
-//}
+inline static void ggml_vec_gelu_quick_f16(const int n, ggml_fp16_t * y, const ggml_fp16_t * x) {
+    const uint16_t * i16 = (const uint16_t *) x;
+    for (int i = 0; i < n; ++i) {
+        y[i] = ggml_table_gelu_quick_f16[i16[i]];
+    }
+}
 
 #ifdef GGML_GELU_QUICK_FP16
 inline static void ggml_vec_gelu_quick_f32(const int n, float * y, const float * x) {
@@ -1060,13 +1060,6 @@ inline static void ggml_vec_gelu_quick_f32(const int n, float * y, const float *
 }
 #endif
 
-inline static void ggml_vec_gelu_quick_f16(const int n, ggml_fp16_t * y, const ggml_fp16_t * x) {
-    for (int i = 0; i < n; ++i) {
-        float v = GGML_CPU_FP16_TO_FP32(x[i]);
-        y[i] = GGML_CPU_FP32_TO_FP16(v*(1.0f/(1.0f+expf(GELU_QUICK_COEF*v))));
-    }
-}
-
 // Sigmoid Linear Unit (SiLU) function
 inline static float ggml_silu_f32(float x) {
     return x/(1.0f + expf(-x));

ggml/src/ggml-cuda/common.cuh

@@ -830,6 +830,18 @@ static __device__ __forceinline__ float ggml_cuda_ue4m3_to_fp32(uint8_t x) {
 #endif // defined(GGML_USE_HIP) && defined(CDNA3) && defined(FP8_AVAILABLE) && HIP_VERSION >= 60200000
 }
 
+static __device__ __forceinline__ uint8_t ggml_cuda_fp32_to_ue4m3(float x) {
+#if defined(BLACKWELL_MMA_AVAILABLE) // This is used for NVFP4 subblock scale quantizations only
+    if (!(x > 0.0f)) {
+        return 0;
+    }
+    const __nv_fp8_e4m3 xf(x);
+    return xf.__x;
+#else
+     NO_DEVICE_CODE; // Used only for NVFP4 Scales for Activations, only for Blackwell
+#endif // defined(BLACKWELL_MMA_AVAILABLE)
+}
+
 __device__ __forceinline__ uint8_t ggml_cuda_float_to_fp4_e2m1(float x, float e) {
     const uint8_t sign_bit = (x < 0.0f) << 3;
     float         ax       = fabsf(x) * e;

ggml/src/ggml-cuda/concat.cu

@@ -1,96 +1,79 @@
 #include "concat.cuh"
 
 // contiguous kernels
-static __global__ void concat_f32_dim0(const float * x, const float * y, float * dst, const int ne0, const int ne00) {
-    int nidx = threadIdx.x + blockIdx.x * blockDim.x;
-    if (nidx >= ne0) {
-        return;
-    }
+template <int dim>
+static __global__ void __launch_bounds__(CUDA_CONCAT_BLOCK_SIZE) concat_f32_cont(const float * x,
+                                                                                 const float * y,
+                                                                                 float *       dst,
+                                                                                 int64_t       ne00,
+                                                                                 int64_t       ne01,
+                                                                                 int64_t       ne02,
+                                                                                 int64_t       ne0,
+                                                                                 int64_t       ne1,
+                                                                                 int64_t       ne2) {
+    static_assert(dim >= 0 && dim <= 2, "dim must be in [0, 2]");
 
-    int offset_dst =
-        nidx +
-        blockIdx.y * ne0 +
-        blockIdx.z * ne0 * gridDim.y;
+    const int64_t n = ne0 * ne1 * ne2;
 
-    if (nidx < ne00) { // src0
-        int offset_src =
-            nidx +
-            blockIdx.y * ne00 +
-            blockIdx.z * ne00 * gridDim.y;
-        dst[offset_dst] = x[offset_src];
-    } else {
-        int offset_src =
-            (nidx - ne00) +
-            blockIdx.y * (ne0 - ne00) +
-            blockIdx.z * (ne0 - ne00) * gridDim.y;
-        dst[offset_dst] = y[offset_src];
+    for (int64_t i = (int64_t) blockIdx.x * blockDim.x + threadIdx.x; i < n; i += (int64_t) blockDim.x * gridDim.x) {
+        if constexpr (dim == 0) {
+            const int64_t row = i / ne0;
+            const int64_t i0  = i - row * ne0;
+
+            if (i0 < ne00) {
+                dst[i] = x[row * ne00 + i0];
+            } else {
+                dst[i] = y[row * (ne0 - ne00) + (i0 - ne00)];
+            }
+        } else if constexpr (dim == 1) {
+            const int64_t dst_plane  = ne0 * ne1;
+            const int64_t src0_plane = ne0 * ne01;
+            const int64_t src1_plane = dst_plane - src0_plane;
+            const int64_t i2         = i / dst_plane;
+            const int64_t i01        = i - i2 * dst_plane;
+
+            if (i01 < src0_plane) {
+                dst[i] = x[i2 * src0_plane + i01];
+            } else {
+                dst[i] = y[i2 * src1_plane + (i01 - src0_plane)];
+            }
+        } else {
+            const int64_t src0_size = ne0 * ne1 * ne02;
+
+            if (i < src0_size) {
+                dst[i] = x[i];
+            } else {
+                dst[i] = y[i - src0_size];
+            }
+        }
     }
 }
 
-static __global__ void concat_f32_dim1(const float * x, const float * y, float * dst, const int ne0, const int ne01) {
-    int nidx = threadIdx.x + blockIdx.x * blockDim.x;
-    if (nidx >= ne0) {
-        return;
-    }
+static void concat_f32_cuda(const float * x,
+                            const float * y,
+                            float *       dst,
+                            int64_t       ne00,
+                            int64_t       ne01,
+                            int64_t       ne02,
+                            int64_t       ne0,
+                            int64_t       ne1,
+                            int64_t       ne2,
+                            int           dim,
+                            cudaStream_t  stream) {
+    const int64_t n          = ne0 * ne1 * ne2;
+    const int     num_blocks = (n + CUDA_CONCAT_BLOCK_SIZE - 1) / CUDA_CONCAT_BLOCK_SIZE;
 
-    int offset_dst =
-        nidx +
-        blockIdx.y * ne0 +
-        blockIdx.z * ne0 * gridDim.y;
-
-    if (blockIdx.y < (unsigned)ne01) { // src0
-        int offset_src =
-            nidx +
-            blockIdx.y * ne0 +
-            blockIdx.z * ne0 * ne01;
-        dst[offset_dst] = x[offset_src];
-    } else {
-        int offset_src =
-            nidx +
-            (blockIdx.y - ne01) * ne0 +
-            blockIdx.z * ne0 * (gridDim.y - ne01);
-        dst[offset_dst] = y[offset_src];
-    }
-}
-
-static __global__ void concat_f32_dim2(const float * x, const float * y, float * dst, const int ne0, const int ne02) {
-    int nidx = threadIdx.x + blockIdx.x * blockDim.x;
-    if (nidx >= ne0) {
-        return;
-    }
-
-    int offset_dst =
-        nidx +
-        blockIdx.y * ne0 +
-        blockIdx.z * ne0 * gridDim.y;
-
-    if (blockIdx.z < (unsigned)ne02) { // src0
-        int offset_src =
-            nidx +
-            blockIdx.y * ne0 +
-            blockIdx.z * ne0 * gridDim.y;
-        dst[offset_dst] = x[offset_src];
-    } else {
-        int offset_src =
-            nidx +
-            blockIdx.y * ne0 +
-            (blockIdx.z - ne02) * ne0 *  gridDim.y;
-        dst[offset_dst] = y[offset_src];
-    }
-}
-
-static void concat_f32_cuda(const float * x, const float * y, float * dst, int ne00, int ne01, int ne02, int ne0, int ne1, int ne2, int dim, cudaStream_t stream) {
-    int num_blocks = (ne0 + CUDA_CONCAT_BLOCK_SIZE - 1) / CUDA_CONCAT_BLOCK_SIZE;
-    dim3 gridDim(num_blocks, ne1, ne2);
     if (dim == 0) {
-        concat_f32_dim0<<<gridDim, CUDA_CONCAT_BLOCK_SIZE, 0, stream>>>(x, y, dst, ne0, ne00);
+        concat_f32_cont<0>
+            <<<num_blocks, CUDA_CONCAT_BLOCK_SIZE, 0, stream>>>(x, y, dst, ne00, ne01, ne02, ne0, ne1, ne2);
         return;
     }
     if (dim == 1) {
-        concat_f32_dim1<<<gridDim, CUDA_CONCAT_BLOCK_SIZE, 0, stream>>>(x, y, dst, ne0, ne01);
+        concat_f32_cont<1>
+            <<<num_blocks, CUDA_CONCAT_BLOCK_SIZE, 0, stream>>>(x, y, dst, ne00, ne01, ne02, ne0, ne1, ne2);
         return;
     }
-    concat_f32_dim2<<<gridDim, CUDA_CONCAT_BLOCK_SIZE, 0, stream>>>(x, y, dst, ne0, ne02);
+    concat_f32_cont<2><<<num_blocks, CUDA_CONCAT_BLOCK_SIZE, 0, stream>>>(x, y, dst, ne00, ne01, ne02, ne0, ne1, ne2);
 }
 
 // non-contiguous kernel (slow)

ggml/src/ggml-cuda/fattn-mma-f16.cuh

@@ -66,6 +66,9 @@ static constexpr __host__ __device__ fattn_mma_config ggml_cuda_fattn_mma_get_co
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(256, 256, 32, 128, 2,  32, 128, 128, 128, 2, true);
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(256, 256, 64, 128, 2,  32, 128, 128, 128, 2, true);
 
+    GGML_CUDA_FATTN_MMA_CONFIG_CASE(320, 256, 32, 128, 2,  32, 128, 128, 128, 1, false);
+    GGML_CUDA_FATTN_MMA_CONFIG_CASE(320, 256, 64, 256, 1,  32, 128, 128, 128, 1, false);
+
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(512, 512,  8,  64, 4,  32, 256, 256, 128, 1, false);
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(512, 512, 16,  64, 4,  32, 256, 256, 128, 1, false);
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(512, 512, 32, 128, 2,  32, 128, 128, 128, 1, false);
@@ -85,6 +88,9 @@ static constexpr __host__ __device__ fattn_mma_config ggml_cuda_fattn_mma_get_co
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(256, 256, 32, 128, 2,  64, 128, 128,  64, 2, true);
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(256, 256, 64, 128, 2,  64, 128, 128,  64, 2, true);
 
+    GGML_CUDA_FATTN_MMA_CONFIG_CASE(320, 256, 32, 128, 2,  32, 128, 128, 128, 1, false);
+    GGML_CUDA_FATTN_MMA_CONFIG_CASE(320, 256, 64, 256, 1,  32, 128, 128, 128, 1, false);
+
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(512, 512,  8,  64, 4,  32,  96,  64, 128, 1, false);
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(512, 512, 16,  64, 4,  32,  96,  64, 128, 1, false);
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(512, 512, 32, 128, 2,  32, 128, 128, 128, 1, false);
@@ -118,6 +124,9 @@ static constexpr __host__ __device__ fattn_mma_config ggml_cuda_fattn_mma_get_co
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(256, 256, 32, 128, 2,  64, 128, 128,  64, 2, true);
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(256, 256, 64, 128, 2,  64, 128, 128,  64, 2, true);
 
+    GGML_CUDA_FATTN_MMA_CONFIG_CASE(320, 256, 32, 128, 2,  64, 160, 128,  64, 2, true);
+    GGML_CUDA_FATTN_MMA_CONFIG_CASE(320, 256, 64, 128, 2,  64, 160, 128,  64, 2, false);
+
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(512, 512, 16,  64, 4,  32, 128, 128, 128, 1, false);
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(512, 512, 32, 128, 2,  32, 128, 128, 128, 1, false);
     GGML_CUDA_FATTN_MMA_CONFIG_CASE(512, 512, 64, 256, 1,  32, 128, 128, 128, 1, false);
@@ -1217,7 +1226,7 @@ static __device__ __forceinline__ void flash_attn_ext_f16_process_tile(
         float KQ_max_scale[cols_per_thread];
 #pragma unroll
         for (int col = 0; col < cols_per_thread; ++col) {
-            const int jc = cols_per_warp == 8 ? T_C_KQ::get_j(col) : T_C_KQ::get_i(2*col);
+            const int jc = (threadIdx.y/np)*cols_per_warp + (cols_per_warp == 8 ? T_C_KQ::get_j(col) : T_C_KQ::get_i(2*col));
             const float sink = sinks_f[jc % ncols2];
 
             const float KQ_max_new = fmaxf(KQ_max[col], sink);
@@ -1825,6 +1834,10 @@ extern DECL_FATTN_MMA_F16_CASE(576, 512, 1, 16);
 extern DECL_FATTN_MMA_F16_CASE(576, 512, 2, 16);
 extern DECL_FATTN_MMA_F16_CASE(576, 512, 4, 16);
 
+// Mistral Small 4 (DKQ=320, DV=256), GQA=32-only build:
+extern DECL_FATTN_MMA_F16_CASE(320, 256,  1, 32);
+extern DECL_FATTN_MMA_F16_CASE(320, 256,  2, 32);
+
 // For GLM 4.7 Flash
 extern DECL_FATTN_MMA_F16_CASE(576, 512,  4,  4);
 extern DECL_FATTN_MMA_F16_CASE(576, 512,  8,  4);

ggml/src/ggml-cuda/fattn-tile.cu

@@ -38,6 +38,10 @@ void ggml_cuda_flash_attn_ext_tile(ggml_backend_cuda_context & ctx, ggml_tensor
             GGML_ASSERT(V->ne[0] == K->ne[0]);
             ggml_cuda_flash_attn_ext_tile_case<256, 256>(ctx, dst);
         } break;
+        case 320: {
+            GGML_ASSERT(V->ne[0] == 256);
+            ggml_cuda_flash_attn_ext_tile_case<320, 256>(ctx, dst);
+        } break;
         case 512: {
             GGML_ASSERT(V->ne[0] == K->ne[0]);
             ggml_cuda_flash_attn_ext_tile_case<512, 512>(ctx, dst);

ggml/src/ggml-cuda/fattn-tile.cuh

@@ -68,6 +68,8 @@ static constexpr __host__ __device__ uint32_t ggml_cuda_fattn_tile_get_config_nv
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(256, 256, 16, 256, 2,  64,  64)
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(256, 256, 32, 256, 2,  64,  64)
 
+    GGML_CUDA_FATTN_TILE_CONFIG_CASE(320, 256, 16, 256, 2,  64,  64)
+
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(512, 512,  4, 128, 2,  64,  64)
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(512, 512,  8, 256, 2,  64,  64)
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(512, 512, 16, 256, 2,  64,  64)
@@ -128,6 +130,8 @@ static constexpr __host__ __device__ uint32_t ggml_cuda_fattn_tile_get_config_nv
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(256, 256, 16, 256, 2,  32, 128)
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(256, 256, 32, 256, 2,  32,  64)
 
+    GGML_CUDA_FATTN_TILE_CONFIG_CASE(320, 256, 16, 256, 2,  32,  64)
+
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(512, 512,  4, 128, 2,  32,  64)
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(512, 512,  8, 256, 2,  32,  64)
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(512, 512, 16, 256, 2,  32,  64)
@@ -195,6 +199,8 @@ static constexpr __host__ __device__ uint32_t ggml_cuda_fattn_tile_get_config_am
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(256, 256, 16, 256, 2,  32, 128)
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(256, 256, 32, 256, 2,  32, 128)
 
+    GGML_CUDA_FATTN_TILE_CONFIG_CASE(320, 256, 32, 512, 1, 128,  64)
+
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(512, 512,  4, 128, 2,  64,  64)
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(512, 512,  8, 256, 2,  64,  64)
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(512, 512, 16, 256, 2,  64,  64)
@@ -264,6 +270,8 @@ static constexpr __host__ __device__ uint32_t ggml_cuda_fattn_tile_get_config_am
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(256, 256, 16, 256, 5,  32, 256)
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(256, 256, 32, 256, 3,  64, 128)
 
+    GGML_CUDA_FATTN_TILE_CONFIG_CASE(320, 256, 32, 256, 2, 128,  64)
+
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(512, 512,  4, 128, 2,  64,  64)
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(512, 512,  8, 256, 2,  64,  64)
     GGML_CUDA_FATTN_TILE_CONFIG_CASE(512, 512, 16, 256, 4,  64,  64)
@@ -1116,7 +1124,7 @@ static void launch_fattn_tile_switch_ncols1(ggml_backend_cuda_context & ctx, ggm
     constexpr size_t nbytes_shared = 0;
 
 #ifdef GGML_USE_HIP
-    if constexpr (DV <= 128) {
+    if constexpr (DKQ <= 128) {
         if (Q->ne[1] > 32/ncols2) {
             constexpr int cols_per_block = 64;
             const int nwarps    = ggml_cuda_fattn_tile_get_nthreads (DKQ, DV, cols_per_block, cc) / warp_size;
@@ -1130,7 +1138,7 @@ static void launch_fattn_tile_switch_ncols1(ggml_backend_cuda_context & ctx, ggm
 #endif // GGML_USE_HIP
 
 #ifndef GGML_USE_HIP
-    if constexpr (DV <= 256)
+    if constexpr (DKQ <= 256)
 #endif // GGML_USE_HIP
     {
         if (Q->ne[1] > 16/ncols2) {
@@ -1144,14 +1152,16 @@ static void launch_fattn_tile_switch_ncols1(ggml_backend_cuda_context & ctx, ggm
         }
     }
 
-    if (Q->ne[1] > 8/ncols2) {
-        constexpr int cols_per_block = 16;
-        const int nwarps    = ggml_cuda_fattn_tile_get_nthreads (DKQ, DV, cols_per_block, cc) / warp_size;
-        const int nbatch_fa = ggml_cuda_fattn_tile_get_nbatch_fa(DKQ, DV, cols_per_block, cc);
-        fattn_kernel_t fattn_kernel = flash_attn_tile<DKQ, DV, cols_per_block/ncols2, ncols2, use_logit_softcap>;
-        launch_fattn<DV, cols_per_block/ncols2, ncols2>
-            (ctx, dst, fattn_kernel, nwarps, nbytes_shared, nbatch_fa, true, true, false, warp_size);
-        return;
+    if constexpr (ncols2 <= 16) {
+        if (Q->ne[1] > 8/ncols2) {
+            constexpr int cols_per_block = 16;
+            const int nwarps    = ggml_cuda_fattn_tile_get_nthreads (DKQ, DV, cols_per_block, cc) / warp_size;
+            const int nbatch_fa = ggml_cuda_fattn_tile_get_nbatch_fa(DKQ, DV, cols_per_block, cc);
+            fattn_kernel_t fattn_kernel = flash_attn_tile<DKQ, DV, cols_per_block/ncols2, ncols2, use_logit_softcap>;
+            launch_fattn<DV, cols_per_block/ncols2, ncols2>
+                (ctx, dst, fattn_kernel, nwarps, nbytes_shared, nbatch_fa, true, true, false, warp_size);
+            return;
+        }
     }
 
     if constexpr (ncols2 <= 8) {
@@ -1210,6 +1220,25 @@ static void launch_fattn_tile_switch_ncols2(ggml_backend_cuda_context & ctx, ggm
     const int gqa_limit = nvidia && gqa_ratio <= 4 && DV <= 256 ? 16 : INT_MAX;
     const bool use_gqa_opt = mask && max_bias == 0.0f && Q->ne[1] <= gqa_limit && K->ne[1] % FATTN_KQ_STRIDE == 0;
 
+    if constexpr (DKQ == 320) {
+        // This branch is only used for Mistral Small 4 which has a GQA ratio of 32.
+        // On AMD, simply use that GQA ratio with 32 columns / block since we always have enough SRAM.
+        // On NVIDIA however, the tile kernel is only used for GPUs that can't use the mma kernel (Pascal and older).
+        // Therefore, use a GQA ratio of 16 with 16 columns / block to stay below 48 kiB of SRAM / block.
+#ifdef GGML_USE_HIP
+        if (use_gqa_opt && gqa_ratio % 32 == 0) {
+            launch_fattn_tile_switch_ncols1<DKQ, DV, 32, use_logit_softcap>(ctx, dst);
+            return;
+        }
+#else
+        if (use_gqa_opt && gqa_ratio % 16 == 0) {
+            launch_fattn_tile_switch_ncols1<DKQ, DV, 16, use_logit_softcap>(ctx, dst);
+            return;
+        }
+#endif // GGML_USE_HIP
+        GGML_ABORT("flash-attn tile (320/256): expected GQA ratio multiple of 32");
+    }
+
     if constexpr (DKQ == 576) {
         if (use_gqa_opt && gqa_ratio % 16 == 0) {
             launch_fattn_tile_switch_ncols1<DKQ, DV, 16, use_logit_softcap>(ctx, dst);
@@ -1221,7 +1250,7 @@ static void launch_fattn_tile_switch_ncols2(ggml_backend_cuda_context & ctx, ggm
         }
     }
 
-    if constexpr (DKQ <= 512) {
+    if constexpr (DKQ <= 512 && DKQ != 320) {
         if (use_gqa_opt && gqa_ratio % 8 == 0) {
             launch_fattn_tile_switch_ncols1<DKQ, DV, 8, use_logit_softcap>(ctx, dst);
             return;
@@ -1275,5 +1304,6 @@ extern DECL_FATTN_TILE_CASE( 96,  96);
 extern DECL_FATTN_TILE_CASE(112, 112);
 extern DECL_FATTN_TILE_CASE(128, 128);
 extern DECL_FATTN_TILE_CASE(256, 256);
+extern DECL_FATTN_TILE_CASE(320, 256);
 extern DECL_FATTN_TILE_CASE(512, 512);
 extern DECL_FATTN_TILE_CASE(576, 512);

ggml/src/ggml-cuda/fattn.cu

@@ -143,6 +143,22 @@ static void ggml_cuda_flash_attn_ext_mma_f16(ggml_backend_cuda_context & ctx, gg
             GGML_ASSERT(V->ne[0] == 256);
             ggml_cuda_flash_attn_ext_mma_f16_switch_ncols2<256, 256>(ctx, dst);
             break;
+        case 320:
+            // For Mistral Small 4, go straight to the ncols1 switch (ncols2=32-only build).
+            GGML_ASSERT(V->ne[0] == 256);
+            {
+                float max_bias = 0.0f;
+                memcpy(&max_bias, (const float *) KQV->op_params + 1, sizeof(float));
+
+                const bool use_gqa_opt = mask && max_bias == 0.0f;
+                GGML_ASSERT(use_gqa_opt);
+                GGML_ASSERT(Q->ne[2] % K->ne[2] == 0);
+                const int gqa_ratio = Q->ne[2] / K->ne[2];
+                GGML_ASSERT(gqa_ratio % 32 == 0);
+
+                ggml_cuda_flash_attn_ext_mma_f16_switch_ncols1<320, 256, 32>(ctx, dst);
+            }
+            break;
         case 512:
             GGML_ASSERT(V->ne[0] == 512);
             ggml_cuda_flash_attn_ext_mma_f16_switch_ncols2<512, 512>(ctx, dst);
@@ -352,6 +368,14 @@ static best_fattn_kernel ggml_cuda_get_best_fattn_kernel(const int device, const
                 return BEST_FATTN_KERNEL_NONE;
             }
             break;
+        case 320:
+            if (V->ne[0] != 256 || !gqa_opt_applies) {
+                return BEST_FATTN_KERNEL_NONE;
+            }
+            if (gqa_ratio % 32 != 0) {
+                return BEST_FATTN_KERNEL_NONE;
+            }
+            break;
         case 512:
             if (V->ne[0] != K->ne[0]) {
                 return BEST_FATTN_KERNEL_NONE;

ggml/src/ggml-cuda/ggml-cuda.cu

@@ -3556,6 +3556,9 @@ static bool ggml_cuda_can_fuse(const struct ggml_cgraph *                cgraph,
      && unary_ops.size() == 1 && unary_ops.begin()[0] == GGML_UNARY_OP_SILU) {
         const ggml_tensor * ssm_conv = cgraph->nodes[node_idx];
         const ggml_tensor * silu     = cgraph->nodes[node_idx+1];
+        if (ggml_get_unary_op(silu) != unary_ops.begin()[0]) {
+            return false;
+        }
 
         if (ssm_conv->type != GGML_TYPE_F32 || silu->type != GGML_TYPE_F32) {
             return false;
@@ -3564,6 +3567,31 @@ static bool ggml_cuda_can_fuse(const struct ggml_cgraph *                cgraph,
         return true;
     }
 
+    if (ops.size() == 3 && ops.begin()[0] == GGML_OP_SSM_CONV && ops.begin()[1] == GGML_OP_ADD
+     && ops.begin()[2] == GGML_OP_UNARY && unary_ops.size() == 1 && unary_ops.begin()[0] == GGML_UNARY_OP_SILU) {
+        const ggml_tensor * ssm_conv = cgraph->nodes[node_idx];
+        const ggml_tensor * add      = cgraph->nodes[node_idx+1];
+        const ggml_tensor * silu     = cgraph->nodes[node_idx+2];
+        if (ggml_get_unary_op(silu) != unary_ops.begin()[0]) {
+            return false;
+        }
+
+        if (ssm_conv->type != GGML_TYPE_F32 || add->type != GGML_TYPE_F32 || silu->type != GGML_TYPE_F32) {
+            return false;
+        }
+
+        // ADD must consume ssm_conv's output and broadcast a 1-D channel-wise bias.
+        const ggml_tensor * bias = (add->src[0] == ssm_conv) ? add->src[1] : add->src[0];
+        if (bias->type != GGML_TYPE_F32 || !ggml_is_contiguous(bias)) {
+            return false;
+        }
+        if (ggml_nelements(bias) != ssm_conv->ne[0] || bias->ne[0] != ssm_conv->ne[0]) {
+            return false;
+        }
+
+        return true;
+    }
+
     if (ops.size() == 2 && ops.begin()[0] == GGML_OP_UNARY && ops.begin()[1] == GGML_OP_MUL
      && unary_ops.size() == 1 && (unary_ops.begin()[0] == GGML_UNARY_OP_SILU || unary_ops.begin()[0] == GGML_UNARY_OP_SIGMOID || unary_ops.begin()[0] == GGML_UNARY_OP_SOFTPLUS)) {
         const ggml_tensor * unary = cgraph->nodes[node_idx];
@@ -3640,6 +3668,362 @@ static bool ggml_cuda_can_fuse(const struct ggml_cgraph *                cgraph,
     return false;
 }
 
+// try and fuse nodes and return the number of nodes to skip
+static int ggml_cuda_try_fuse(ggml_backend_cuda_context * cuda_ctx, ggml_cgraph * cgraph, int i) {
+
+    static bool disable_fusion = getenv("GGML_CUDA_DISABLE_FUSION") != nullptr && std::atoi(getenv("GGML_CUDA_DISABLE_FUSION"));
+    if (disable_fusion) {
+        return 0;
+    }
+
+    ggml_tensor * node = cgraph->nodes[i];
+
+    //topk-moe
+    if (cgraph->nodes[i]->op == GGML_OP_UNARY || cgraph->nodes[i]->op == GGML_OP_SOFT_MAX ||
+            cgraph->nodes[i]->op == GGML_OP_ARGSORT) {
+        ggml_cuda_topk_moe_args args;
+        const bool              can_fuse = ggml_cuda_topk_moe_fusion(cgraph, i, args);
+        std::vector<ggml_op>    ops;
+
+        if (can_fuse) {
+            const ggml_tensor * logits  = node->src[0];
+            ggml_tensor *       weights = nullptr;
+            ggml_tensor *       ids     = nullptr;
+            const ggml_tensor * bias    = nullptr;
+            const ggml_tensor * clamp   = nullptr;
+            const ggml_tensor * scale   = nullptr;
+
+            if (!args.delayed_softmax) {
+                ggml_op gating_op = args.sigmoid ? GGML_OP_UNARY : GGML_OP_SOFT_MAX;
+                int     out_nodes[2];  // nodes which can't be elided
+
+                if (args.prob_bias) {
+                    bias = cgraph->nodes[i + 2]->src[1];
+                    ops.insert(ops.end(), { gating_op, GGML_OP_RESHAPE, GGML_OP_ADD, GGML_OP_ARGSORT, GGML_OP_VIEW,
+                                            GGML_OP_GET_ROWS });
+                    out_nodes[0] = i + 4;
+                    ids          = cgraph->nodes[i + 4];
+                } else {
+                    ops.insert(ops.end(),
+                               { gating_op, GGML_OP_RESHAPE, GGML_OP_ARGSORT, GGML_OP_VIEW, GGML_OP_GET_ROWS });
+                    out_nodes[0] = i + 3;
+                    ids          = cgraph->nodes[i + 3];
+                }
+
+                if (args.norm) {
+                    ops.insert(ops.end(),
+                               { GGML_OP_RESHAPE, GGML_OP_SUM_ROWS, GGML_OP_CLAMP, GGML_OP_DIV, GGML_OP_RESHAPE });
+                    clamp = cgraph->nodes[i + ops.size() - 3];
+                }
+                if (args.scale) {
+                    ops.insert(ops.end(), { GGML_OP_SCALE });
+                    scale = cgraph->nodes[i + ops.size() - 1];
+                }
+
+                weights      = cgraph->nodes[i + ops.size() - 1];
+                out_nodes[1] = i + ops.size() - 1;
+
+                if (ggml_can_fuse_subgraph(cgraph, i, ops.size(), ops.data(), out_nodes, 2) &&
+                        ggml_cuda_should_use_topk_moe(node, logits, weights, ids) &&
+                        ggml_cuda_check_fusion_memory_ranges(cgraph, i, ops.size(), out_nodes, 2, /*is_topk_moe=*/true)) {
+                    ggml_cuda_op_topk_moe(*cuda_ctx, logits, weights, ids, clamp, scale, bias, args);
+                    return ops.size() - 1;
+                }
+            } else if (!args.norm && !args.prob_bias) {
+                //special case gpt-oss, no norm, no bias.
+                ops.insert(ops.end(), { GGML_OP_ARGSORT, GGML_OP_VIEW, GGML_OP_GET_ROWS, GGML_OP_RESHAPE,
+                                        GGML_OP_SOFT_MAX, GGML_OP_RESHAPE });
+                weights                     = cgraph->nodes[i + 5];
+                ids                         = cgraph->nodes[i + 1];
+                const ggml_tensor * softmax = cgraph->nodes[i + 4];
+
+                int out_nodes[2] = { i + 1, i + 5 };
+                if (ggml_can_fuse_subgraph(cgraph, i, ops.size(), ops.data(), out_nodes, 2) &&
+                        ggml_cuda_should_use_topk_moe(softmax, logits, weights, ids) &&
+                        ggml_cuda_check_fusion_memory_ranges(cgraph, i, ops.size(), out_nodes, 2, /*is_topk_moe=*/true)) {
+                    ggml_cuda_op_topk_moe(*cuda_ctx, logits, weights, ids, clamp, scale, bias, args);
+                    return ops.size() - 1;
+                }
+            }
+        }
+    }
+
+    //RoPE + view + set-rows
+    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_ROPE, GGML_OP_VIEW, GGML_OP_SET_ROWS }, {})) {
+        ggml_tensor * rope     = cgraph->nodes[i];
+        ggml_tensor * set_rows = cgraph->nodes[i + 2];
+
+        ggml_cuda_op_rope_fused(*cuda_ctx, rope, set_rows);
+        return 2;
+    }
+
+    // multi-(add or mul)
+    if (node->op == GGML_OP_ADD || node->op == GGML_OP_MUL) {
+        int     n_fuse = 0;
+        ggml_op ops[8];
+        std::fill(ops, ops + 8, node->op);
+
+        for (; n_fuse <= 6; ++n_fuse) {
+            if (!ggml_can_fuse(cgraph, i + n_fuse, ops + n_fuse, 2)) {
+                break;
+            }
+            if (cgraph->nodes[i + n_fuse] != cgraph->nodes[i + n_fuse + 1]->src[0]) {
+                break;
+            }
+            if (!ggml_are_same_layout(cgraph->nodes[i + n_fuse]->src[1], cgraph->nodes[i + n_fuse + 1]->src[1])) {
+                break;
+            }
+        }
+
+        n_fuse++;
+
+        if (n_fuse > 1) {
+            ggml_tensor fused_node;
+            memcpy(&fused_node, node, sizeof(ggml_tensor));
+            for (int j = 0; j < n_fuse - 1; ++j) {
+                fused_node.src[j + 2] = cgraph->nodes[i + j + 1]->src[1];
+            }
+            fused_node.data = cgraph->nodes[i + n_fuse - 1]->data;
+            if (node->op == GGML_OP_ADD) {
+                ggml_cuda_op_fused_add(*cuda_ctx, &fused_node, n_fuse);
+            } else {
+                ggml_cuda_op_fused_mul(*cuda_ctx, &fused_node, n_fuse);
+            }
+            return n_fuse - 1;
+        }
+    }
+
+    bool fused_mul_mat_vec = false;
+    int  fused_node_count  = 0;
+
+    // gate + glu + up
+    for (ggml_op op : { GGML_OP_MUL_MAT, GGML_OP_MUL_MAT_ID }) {
+        const ggml_op bias_op = op == GGML_OP_MUL_MAT ? GGML_OP_ADD : GGML_OP_ADD_ID;
+
+        if (ggml_cuda_can_fuse(cgraph, i, { op, bias_op, op, bias_op, GGML_OP_GLU }, {})) {
+            ggml_tensor * glu         = cgraph->nodes[i + 4];
+            ggml_tensor * gate_bias_n = glu->src[0];
+            ggml_tensor * up_bias_n   = glu->src[1];
+
+            //we don't assume the order for {gate, up}. Instead infer it from the bias tensor
+            ggml_tensor * gate_n = nullptr;
+            ggml_tensor * up_n   = nullptr;
+
+            if (gate_bias_n->src[0] == cgraph->nodes[i] || gate_bias_n->src[1] == cgraph->nodes[i]) {
+                gate_n = cgraph->nodes[i];
+                up_n   = cgraph->nodes[i + 2];
+            } else if (gate_bias_n->src[0] == cgraph->nodes[i + 2] || gate_bias_n->src[1] == cgraph->nodes[i + 2]) {
+                gate_n = cgraph->nodes[i + 2];
+                up_n   = cgraph->nodes[i];
+            } else {
+                continue;
+            }
+
+            auto get_bias_tensor = [](const ggml_tensor * bias_node, const ggml_tensor * mul_node, ggml_op op_bias) {
+                if (op_bias == GGML_OP_ADD) {
+                    if (bias_node->src[0] == mul_node) {
+                        return bias_node->src[1];
+                    }
+                    if (bias_node->src[1] == mul_node) {
+                        return bias_node->src[0];
+                    }
+                    return (ggml_tensor *) nullptr;
+                }
+                GGML_ASSERT(op_bias == GGML_OP_ADD_ID);
+                GGML_ASSERT(bias_node->src[0] == mul_node);
+                return bias_node->src[1];
+            };
+
+            ggml_tensor * up_bias_tensor   = get_bias_tensor(up_bias_n, up_n, bias_op);
+            ggml_tensor * gate_bias_tensor = get_bias_tensor(gate_bias_n, gate_n, bias_op);
+
+            if (!up_bias_tensor || !gate_bias_tensor) {
+                continue;
+            }
+
+            // we don't support repeating adds
+            if (bias_op == GGML_OP_ADD && (!ggml_are_same_shape(gate_bias_n->src[0], gate_bias_n->src[1]) ||
+                                           !ggml_are_same_shape(up_bias_n->src[0], up_bias_n->src[1]))) {
+                continue;
+            }
+
+            const ggml_tensor * src0 = up_n->src[0];
+            const ggml_tensor * src1 = up_n->src[1];
+            const ggml_tensor * ids  = up_n->src[2];
+
+            if (ggml_cuda_should_fuse_mul_mat_vec_f(up_n)) {
+                ggml_cuda_mm_fusion_args_host fusion_data{};
+                fusion_data.gate      = gate_n->src[0];
+                fusion_data.x_bias    = up_bias_tensor;
+                fusion_data.gate_bias = gate_bias_tensor;
+                fusion_data.glu_op    = ggml_get_glu_op(glu);
+
+                ggml_cuda_mul_mat_vec_f(*cuda_ctx, src0, src1, ids, glu, &fusion_data);
+                fused_mul_mat_vec = true;
+                fused_node_count  = 5;
+                break;
+            }
+
+            if (ggml_cuda_should_fuse_mul_mat_vec_q(up_n)) {
+                ggml_cuda_mm_fusion_args_host fusion_data{};
+                fusion_data.gate      = gate_n->src[0];
+                fusion_data.x_bias    = up_bias_tensor;
+                fusion_data.gate_bias = gate_bias_tensor;
+                fusion_data.glu_op    = ggml_get_glu_op(glu);
+
+                ggml_cuda_mul_mat_vec_q(*cuda_ctx, src0, src1, ids, glu, &fusion_data);
+                fused_mul_mat_vec = true;
+                fused_node_count  = 5;
+                break;
+            }
+        } else if (ggml_cuda_can_fuse(cgraph, i, { op, op, GGML_OP_GLU }, {})) {
+            ggml_tensor * glu  = cgraph->nodes[i + 2];
+            ggml_tensor * gate = glu->src[0];
+            ggml_tensor * up   = glu->src[1];
+
+            bool ok = (gate == cgraph->nodes[i] && up == cgraph->nodes[i + 1]) ||
+                      (gate == cgraph->nodes[i + 1] && up == cgraph->nodes[i]);
+
+            if (!ok) {
+                continue;
+            }
+
+            const ggml_tensor * src0 = up->src[0];
+            const ggml_tensor * src1 = up->src[1];
+            const ggml_tensor * ids  = up->src[2];
+
+            if (ggml_cuda_should_fuse_mul_mat_vec_f(up)) {
+                ggml_cuda_mm_fusion_args_host fusion_data{};
+                fusion_data.gate   = gate->src[0];
+                fusion_data.glu_op = ggml_get_glu_op(glu);
+
+                ggml_cuda_mul_mat_vec_f(*cuda_ctx, src0, src1, ids, glu, &fusion_data);
+                fused_mul_mat_vec = true;
+                fused_node_count  = 3;
+                break;
+            }
+
+            if (ggml_cuda_should_fuse_mul_mat_vec_q(up)) {
+                ggml_cuda_mm_fusion_args_host fusion_data{};
+                fusion_data.gate   = gate->src[0];
+                fusion_data.glu_op = ggml_get_glu_op(glu);
+
+                ggml_cuda_mul_mat_vec_q(*cuda_ctx, src0, src1, ids, glu, &fusion_data);
+                fused_mul_mat_vec = true;
+                fused_node_count  = 3;
+                break;
+            }
+        }
+    }
+
+    if (fused_mul_mat_vec) {
+        return fused_node_count - 1;
+    }
+
+    fused_mul_mat_vec = false;
+    fused_node_count  = 0;
+
+    // gate + add + glu + up + add
+    for (ggml_op op : { GGML_OP_MUL_MAT, GGML_OP_MUL_MAT_ID }) {
+        const ggml_op bias_op = op == GGML_OP_MUL_MAT ? GGML_OP_ADD : GGML_OP_ADD_ID;
+
+        if (!ggml_can_fuse(cgraph, i, { op, bias_op })) {
+            continue;
+        }
+
+        ggml_tensor * mm_node   = cgraph->nodes[i];
+        ggml_tensor * bias_node = cgraph->nodes[i + 1];
+
+        ggml_tensor * bias_tensor = nullptr;
+        if (bias_op == GGML_OP_ADD) {
+            if (bias_node->src[0] == mm_node) {
+                bias_tensor = bias_node->src[1];
+            } else if (bias_node->src[1] == mm_node) {
+                bias_tensor = bias_node->src[0];
+            } else {
+                continue;
+            }
+        } else {
+            if (bias_node->src[0] != mm_node) {
+                continue;
+            }
+            bias_tensor = bias_node->src[1];
+        }
+
+        const ggml_tensor * src0 = mm_node->src[0];
+        const ggml_tensor * src1 = mm_node->src[1];
+        const ggml_tensor * ids  = mm_node->src[2];
+
+        if (bias_op == GGML_OP_ADD_ID && bias_node->src[2] != ids) {
+            continue;
+        }
+
+        if (bias_op == GGML_OP_ADD && !ggml_are_same_shape(bias_node->src[0], bias_node->src[1])) {
+            continue;
+        }
+
+        ggml_cuda_mm_fusion_args_host fusion_data{};
+        fusion_data.x_bias = bias_tensor;
+
+        if (ggml_cuda_should_fuse_mul_mat_vec_f(mm_node)) {
+            ggml_cuda_mul_mat_vec_f(*cuda_ctx, src0, src1, ids, bias_node, &fusion_data);
+            fused_mul_mat_vec = true;
+            fused_node_count  = 2;
+            break;
+        }
+
+        if (ggml_cuda_should_fuse_mul_mat_vec_q(mm_node)) {
+            ggml_cuda_mul_mat_vec_q(*cuda_ctx, src0, src1, ids, bias_node, &fusion_data);
+            fused_mul_mat_vec = true;
+            fused_node_count  = 2;
+            break;
+        }
+    }
+
+    if (fused_mul_mat_vec) {
+        return fused_node_count - 1;
+    }
+
+    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_RMS_NORM, GGML_OP_MUL, GGML_OP_ADD }, {})) {
+        ggml_cuda_op_rms_norm_fused_add(*cuda_ctx, node, cgraph->nodes[i + 1], cgraph->nodes[i + 2]);
+        return 2;
+    }
+
+    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_RMS_NORM, GGML_OP_MUL }, {})) {
+        ggml_cuda_op_rms_norm_fused(*cuda_ctx, node, cgraph->nodes[i + 1]);
+        return 1;
+    }
+
+    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_SSM_CONV, GGML_OP_ADD, GGML_OP_UNARY }, { GGML_UNARY_OP_SILU })) {
+        ggml_cuda_op_ssm_conv(*cuda_ctx, node, cgraph->nodes[i + 1], cgraph->nodes[i + 2]);
+        return 2;
+    }
+
+    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_SSM_CONV, GGML_OP_UNARY }, { GGML_UNARY_OP_SILU })) {
+        ggml_cuda_op_ssm_conv(*cuda_ctx, node, /*bias_add_node=*/ nullptr, cgraph->nodes[i + 1]);
+        return 1;
+    }
+
+    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_UNARY, GGML_OP_MUL }, { GGML_UNARY_OP_SILU }) ||
+        ggml_cuda_can_fuse(cgraph, i, { GGML_OP_UNARY, GGML_OP_MUL }, { GGML_UNARY_OP_SIGMOID }) ||
+        ggml_cuda_can_fuse(cgraph, i, { GGML_OP_UNARY, GGML_OP_MUL }, { GGML_UNARY_OP_SOFTPLUS })) {
+        ggml_cuda_op_unary_mul(*cuda_ctx, node, cgraph->nodes[i + 1]);
+        return 1;
+    }
+
+    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_UNARY, GGML_OP_SQR }, { GGML_UNARY_OP_RELU })) {
+        ggml_cuda_op_relu_sqr(*cuda_ctx, node, cgraph->nodes[i + 1]);
+        return 1;
+    }
+
+    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_SCALE, GGML_OP_UNARY, GGML_OP_SCALE }, { GGML_UNARY_OP_TANH })) {
+        ggml_cuda_op_softcap(*cuda_ctx, cgraph->nodes[i + 2], node);
+        return 2;
+    }
+
+    return 0;
+}
+
 static void ggml_cuda_graph_evaluate_and_capture(ggml_backend_cuda_context * cuda_ctx, ggml_cgraph * cgraph, const bool use_cuda_graph, const bool cuda_graph_update_required, const void * graph_key) {
     bool graph_evaluated_or_captured = false;
 
@@ -3786,355 +4170,11 @@ static void ggml_cuda_graph_evaluate_and_capture(ggml_backend_cuda_context * cud
                     continue;
                 }
 
-                // start of fusion operations
-                static bool disable_fusion = (getenv("GGML_CUDA_DISABLE_FUSION") != nullptr);
-                if (!disable_fusion) {
-                    ggml_cuda_topk_moe_args args;
+                int nodes_to_skip = ggml_cuda_try_fuse(cuda_ctx, cgraph, i);
 
-                    if (cgraph->nodes[i]->op == GGML_OP_UNARY || cgraph->nodes[i]->op == GGML_OP_SOFT_MAX ||
-                        cgraph->nodes[i]->op == GGML_OP_ARGSORT) {
-                        const bool can_fuse = ggml_cuda_topk_moe_fusion(cgraph, i, args);
-
-                        std::vector<ggml_op> ops;
-
-                        if (can_fuse) {
-                            const ggml_tensor * logits  = node->src[0];
-                            ggml_tensor *       weights = nullptr;
-                            ggml_tensor *       ids     = nullptr;
-                            const ggml_tensor * bias    = nullptr;
-                            const ggml_tensor * clamp   = nullptr;
-                            const ggml_tensor * scale   = nullptr;
-
-                            if (!args.delayed_softmax) {
-                                ggml_op gating_op = args.sigmoid ? GGML_OP_UNARY : GGML_OP_SOFT_MAX;
-                                int     out_nodes[2];  // nodes which can't be elided
-
-                                if (args.prob_bias) {
-                                    bias = cgraph->nodes[i + 2]->src[1];
-                                    ops.insert(ops.end(), { gating_op, GGML_OP_RESHAPE, GGML_OP_ADD, GGML_OP_ARGSORT,
-                                                            GGML_OP_VIEW, GGML_OP_GET_ROWS });
-                                    out_nodes[0] = i + 4;
-                                    ids          = cgraph->nodes[i + 4];
-                                } else {
-                                    ops.insert(ops.end(), { gating_op, GGML_OP_RESHAPE, GGML_OP_ARGSORT, GGML_OP_VIEW,
-                                                            GGML_OP_GET_ROWS });
-                                    out_nodes[0] = i + 3;
-                                    ids          = cgraph->nodes[i + 3];
-                                }
-
-                                if (args.norm) {
-                                    ops.insert(ops.end(), { GGML_OP_RESHAPE, GGML_OP_SUM_ROWS, GGML_OP_CLAMP,
-                                                            GGML_OP_DIV, GGML_OP_RESHAPE });
-                                    clamp = cgraph->nodes[i + ops.size() - 3];
-                                }
-                                if (args.scale) {
-                                    ops.insert(ops.end(), { GGML_OP_SCALE });
-                                    scale = cgraph->nodes[i + ops.size() - 1];
-                                }
-
-                                weights      = cgraph->nodes[i + ops.size() - 1];
-                                out_nodes[1] = i + ops.size() - 1;
-
-                                if (ggml_can_fuse_subgraph(cgraph, i, ops.size(), ops.data(), out_nodes, 2) &&
-                                    ggml_cuda_should_use_topk_moe(node, logits, weights, ids) &&
-                                    ggml_cuda_check_fusion_memory_ranges(cgraph, i, ops.size(), out_nodes, 2, /*is_topk_moe=*/ true)) {
-                                    ggml_cuda_op_topk_moe(*cuda_ctx, logits, weights, ids, clamp, scale, bias, args);
-                                    i += ops.size() - 1;
-                                    continue;
-                                }
-                            } else if (!args.norm && !args.prob_bias) {
-                                //special case gpt-oss, no norm, no bias.
-                                ops.insert(ops.end(), { GGML_OP_ARGSORT, GGML_OP_VIEW, GGML_OP_GET_ROWS,
-                                                        GGML_OP_RESHAPE, GGML_OP_SOFT_MAX, GGML_OP_RESHAPE });
-                                weights                     = cgraph->nodes[i + 5];
-                                ids                         = cgraph->nodes[i + 1];
-                                const ggml_tensor * softmax = cgraph->nodes[i + 4];
-
-                                int out_nodes[2] = { i + 1, i + 5 };
-                                if (ggml_can_fuse_subgraph(cgraph, i, ops.size(), ops.data(), out_nodes, 2) &&
-                                    ggml_cuda_should_use_topk_moe(softmax, logits, weights, ids) &&
-                                    ggml_cuda_check_fusion_memory_ranges(cgraph, i, ops.size(), out_nodes, 2, /*is_topk_moe=*/ true)) {
-                                    ggml_cuda_op_topk_moe(*cuda_ctx, logits, weights, ids, clamp, scale, bias, args);
-                                    i += ops.size() - 1;
-                                    continue;
-                                }
-                            }
-                        }
-                    }
-
-                    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_ROPE, GGML_OP_VIEW, GGML_OP_SET_ROWS }, {})) {
-                        ggml_tensor * rope = cgraph->nodes[i];
-                        ggml_tensor * set_rows = cgraph->nodes[i + 2];
-
-                        ggml_cuda_op_rope_fused(*cuda_ctx, rope, set_rows);
-                        i += 2;
-                        continue;
-                    }
-
-                    if (node->op == GGML_OP_ADD || node->op == GGML_OP_MUL) {
-                        int n_fuse = 0;
-                        ggml_op ops[8];
-                        std::fill(ops, ops + 8, node->op);
-
-                        for (; n_fuse <= 6; ++n_fuse){
-                            if (!ggml_can_fuse(cgraph, i + n_fuse, ops + n_fuse, 2)) {
-                                break;
-                            }
-                            if (cgraph->nodes[i + n_fuse] != cgraph->nodes[i + n_fuse + 1]->src[0]) {
-                                break;
-                            }
-                            if (!ggml_are_same_layout(cgraph->nodes[i + n_fuse]->src[1], cgraph->nodes[i + n_fuse + 1]->src[1])) {
-                                break;
-                            }
-                        }
-
-                        n_fuse++;
-
-                        if (n_fuse > 1) {
-                            ggml_tensor fused_node;
-                            memcpy(&fused_node, node, sizeof(ggml_tensor));
-                            for (int j = 0; j < n_fuse - 1; ++j) {
-                                fused_node.src[j + 2] = cgraph->nodes[i + j + 1]->src[1];
-                            }
-                            fused_node.data = cgraph->nodes[i + n_fuse - 1]->data;
-                            if (node->op == GGML_OP_ADD) {
-                                ggml_cuda_op_fused_add(*cuda_ctx, &fused_node, n_fuse);
-                            } else {
-                                ggml_cuda_op_fused_mul(*cuda_ctx, &fused_node, n_fuse);
-                            }
-                            i += n_fuse - 1;
-
-                            continue;
-                        }
-                    }
-
-                    bool fused_mul_mat_vec = false;
-                    int fused_node_count = 0;
-
-                    for (ggml_op op : { GGML_OP_MUL_MAT, GGML_OP_MUL_MAT_ID }) {
-                        const ggml_op bias_op = op == GGML_OP_MUL_MAT ? GGML_OP_ADD : GGML_OP_ADD_ID;
-
-                        if (ggml_cuda_can_fuse(cgraph, i, { op, bias_op, op, bias_op, GGML_OP_GLU }, {})) {
-                            ggml_tensor * glu         = cgraph->nodes[i + 4];
-                            ggml_tensor * gate_bias_n = glu->src[0];
-                            ggml_tensor * up_bias_n   = glu->src[1];
-
-                            //we don't assume the order for {gate, up}. Instead infer it from the bias tensor
-                            ggml_tensor * gate_n      = nullptr;
-                            ggml_tensor * up_n        = nullptr;
-
-                            if (gate_bias_n->src[0] == cgraph->nodes[i] || gate_bias_n->src[1] == cgraph->nodes[i]) {
-                                gate_n = cgraph->nodes[i];
-                                up_n   = cgraph->nodes[i + 2];
-                            } else if (gate_bias_n->src[0] == cgraph->nodes[i + 2] || gate_bias_n->src[1] == cgraph->nodes[i + 2]) {
-                                gate_n = cgraph->nodes[i + 2];
-                                up_n   = cgraph->nodes[i];
-                            } else {
-                                continue;
-                            }
-
-                            auto get_bias_tensor = [](const ggml_tensor * bias_node, const ggml_tensor * mul_node, ggml_op op_bias) {
-                                if (op_bias == GGML_OP_ADD) {
-                                    if (bias_node->src[0] == mul_node) {
-                                        return bias_node->src[1];
-                                    }
-                                    if (bias_node->src[1] == mul_node) {
-                                        return bias_node->src[0];
-                                    }
-                                    return (ggml_tensor *) nullptr;
-                                }
-                                GGML_ASSERT(op_bias == GGML_OP_ADD_ID);
-                                GGML_ASSERT(bias_node->src[0] == mul_node);
-                                return bias_node->src[1];
-                            };
-
-                            ggml_tensor * up_bias_tensor   = get_bias_tensor(up_bias_n, up_n, bias_op);
-                            ggml_tensor * gate_bias_tensor = get_bias_tensor(gate_bias_n, gate_n, bias_op);
-
-                            if (!up_bias_tensor || !gate_bias_tensor) {
-                                continue;
-                            }
-
-                            // we don't support repeating adds
-                            if (bias_op == GGML_OP_ADD &&
-                                (!ggml_are_same_shape(gate_bias_n->src[0], gate_bias_n->src[1]) ||
-                                 !ggml_are_same_shape(up_bias_n->src[0], up_bias_n->src[1]))) {
-                                continue;
-                            }
-
-                            const ggml_tensor * src0 = up_n->src[0];
-                            const ggml_tensor * src1 = up_n->src[1];
-                            const ggml_tensor * ids  = up_n->src[2];
-
-                            if (ggml_cuda_should_fuse_mul_mat_vec_f(up_n)) {
-                                ggml_cuda_mm_fusion_args_host fusion_data{};
-                                fusion_data.gate      = gate_n->src[0];
-                                fusion_data.x_bias    = up_bias_tensor;
-                                fusion_data.gate_bias = gate_bias_tensor;
-                                fusion_data.glu_op    = ggml_get_glu_op(glu);
-
-                                ggml_cuda_mul_mat_vec_f(*cuda_ctx, src0, src1, ids, glu, &fusion_data);
-                                fused_mul_mat_vec = true;
-                                fused_node_count = 5;
-                                break;
-                            }
-
-                            if (ggml_cuda_should_fuse_mul_mat_vec_q(up_n)) {
-                                ggml_cuda_mm_fusion_args_host fusion_data{};
-                                fusion_data.gate      = gate_n->src[0];
-                                fusion_data.x_bias    = up_bias_tensor;
-                                fusion_data.gate_bias = gate_bias_tensor;
-                                fusion_data.glu_op    = ggml_get_glu_op(glu);
-
-                                ggml_cuda_mul_mat_vec_q(*cuda_ctx, src0, src1, ids, glu, &fusion_data);
-                                fused_mul_mat_vec = true;
-                                fused_node_count = 5;
-                                break;
-                            }
-                        } else if (ggml_cuda_can_fuse(cgraph, i, { op, op, GGML_OP_GLU }, {})) {
-                            ggml_tensor * glu  = cgraph->nodes[i + 2];
-                            ggml_tensor * gate = glu->src[0];
-                            ggml_tensor * up   = glu->src[1];
-
-                            bool ok = (gate == cgraph->nodes[i] && up == cgraph->nodes[i + 1])
-                                || (gate == cgraph->nodes[i + 1] && up == cgraph->nodes[i]);
-
-                            if (!ok) continue;
-
-                            const ggml_tensor * src0 = up->src[0];
-                            const ggml_tensor * src1 = up->src[1];
-                            const ggml_tensor * ids  = up->src[2];
-
-                            if (ggml_cuda_should_fuse_mul_mat_vec_f(up)) {
-                                ggml_cuda_mm_fusion_args_host fusion_data{};
-                                fusion_data.gate   = gate->src[0];
-                                fusion_data.glu_op = ggml_get_glu_op(glu);
-
-                                ggml_cuda_mul_mat_vec_f(*cuda_ctx, src0, src1, ids, glu, &fusion_data);
-                                fused_mul_mat_vec = true;
-                                fused_node_count = 3;
-                                break;
-                            }
-
-                            if (ggml_cuda_should_fuse_mul_mat_vec_q(up)) {
-                                ggml_cuda_mm_fusion_args_host fusion_data{};
-                                fusion_data.gate   = gate->src[0];
-                                fusion_data.glu_op = ggml_get_glu_op(glu);
-
-                                ggml_cuda_mul_mat_vec_q(*cuda_ctx, src0, src1, ids, glu, &fusion_data);
-                                fused_mul_mat_vec = true;
-                                fused_node_count = 3;
-                                break;
-                            }
-                        }
-                    }
-
-                    if (fused_mul_mat_vec) {
-                        i += fused_node_count - 1;
-                        continue;
-                    }
-
-                    fused_mul_mat_vec = false;
-                    fused_node_count = 0;
-
-                    for (ggml_op op : { GGML_OP_MUL_MAT, GGML_OP_MUL_MAT_ID }) {
-                        const ggml_op bias_op = op == GGML_OP_MUL_MAT ? GGML_OP_ADD : GGML_OP_ADD_ID;
-
-                        if (!ggml_can_fuse(cgraph, i, { op, bias_op })) {
-                            continue;
-                        }
-
-                        ggml_tensor * mm_node   = cgraph->nodes[i];
-                        ggml_tensor * bias_node = cgraph->nodes[i + 1];
-
-                        ggml_tensor * bias_tensor = nullptr;
-                        if (bias_op == GGML_OP_ADD) {
-                            if (bias_node->src[0] == mm_node) {
-                                bias_tensor = bias_node->src[1];
-                            } else if (bias_node->src[1] == mm_node) {
-                                bias_tensor = bias_node->src[0];
-                            } else {
-                                continue;
-                            }
-                        } else {
-                            if (bias_node->src[0] != mm_node) {
-                                continue;
-                            }
-                            bias_tensor = bias_node->src[1];
-                        }
-
-                        const ggml_tensor * src0 = mm_node->src[0];
-                        const ggml_tensor * src1 = mm_node->src[1];
-                        const ggml_tensor * ids  = mm_node->src[2];
-
-                        if (bias_op == GGML_OP_ADD_ID && bias_node->src[2] != ids) {
-                            continue;
-                        }
-
-                        if (bias_op == GGML_OP_ADD && !ggml_are_same_shape(bias_node->src[0], bias_node->src[1])) {
-                            continue;
-                        }
-
-                        ggml_cuda_mm_fusion_args_host fusion_data{};
-                        fusion_data.x_bias = bias_tensor;
-
-                        if (ggml_cuda_should_fuse_mul_mat_vec_f(mm_node)) {
-                            ggml_cuda_mul_mat_vec_f(*cuda_ctx, src0, src1, ids, bias_node, &fusion_data);
-                            fused_mul_mat_vec = true;
-                            fused_node_count = 2;
-                            break;
-                        }
-
-                        if (ggml_cuda_should_fuse_mul_mat_vec_q(mm_node)) {
-                            ggml_cuda_mul_mat_vec_q(*cuda_ctx, src0, src1, ids, bias_node, &fusion_data);
-                            fused_mul_mat_vec = true;
-                            fused_node_count = 2;
-                            break;
-                        }
-                    }
-
-                    if (fused_mul_mat_vec) {
-                        i += fused_node_count - 1;
-                        continue;
-                    }
-
-                    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_RMS_NORM, GGML_OP_MUL, GGML_OP_ADD}, {})) {
-                        ggml_cuda_op_rms_norm_fused_add(*cuda_ctx, node, cgraph->nodes[i+1], cgraph->nodes[i+2]);
-                        i += 2;
-                        continue;
-                    }
-
-                    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_RMS_NORM, GGML_OP_MUL}, {})) {
-                        ggml_cuda_op_rms_norm_fused(*cuda_ctx, node, cgraph->nodes[i+1]);
-                        i++;
-                        continue;
-                    }
-
-                    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_SSM_CONV, GGML_OP_UNARY }, { GGML_UNARY_OP_SILU })) {
-                        ggml_cuda_op_ssm_conv(*cuda_ctx, node, cgraph->nodes[i+1]);
-                        i++;
-                        continue;
-                    }
-
-                    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_UNARY, GGML_OP_MUL }, { GGML_UNARY_OP_SILU }) ||
-                        ggml_cuda_can_fuse(cgraph, i, { GGML_OP_UNARY, GGML_OP_MUL }, { GGML_UNARY_OP_SIGMOID }) ||
-                        ggml_cuda_can_fuse(cgraph, i, { GGML_OP_UNARY, GGML_OP_MUL }, { GGML_UNARY_OP_SOFTPLUS })) {
-                        ggml_cuda_op_unary_mul(*cuda_ctx, node, cgraph->nodes[i+1]);
-                        i++;
-                        continue;
-                    }
-
-                    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_UNARY, GGML_OP_SQR }, { GGML_UNARY_OP_RELU })) {
-                        ggml_cuda_op_relu_sqr(*cuda_ctx, node, cgraph->nodes[i+1]);
-                        i++;
-                        continue;
-                    }
-
-                    if (ggml_cuda_can_fuse(cgraph, i, { GGML_OP_SCALE, GGML_OP_UNARY, GGML_OP_SCALE }, { GGML_UNARY_OP_TANH })) {
-                        i += 2;
-                        ggml_cuda_op_softcap(*cuda_ctx, cgraph->nodes[i], node);
-                        continue;
-                    }
+                if (nodes_to_skip != 0) {
+                    i += nodes_to_skip;
+                    continue;
                 }
 #ifndef NDEBUG
                 assert(node->buffer->buft == ggml_backend_cuda_buffer_type(cuda_ctx->device));
@@ -4548,8 +4588,8 @@ static const ggml_backend_i ggml_backend_cuda_interface = {
     /* .free                    = */ ggml_backend_cuda_free,
     /* .set_tensor_async        = */ ggml_backend_cuda_set_tensor_async,
     /* .get_tensor_async        = */ ggml_backend_cuda_get_tensor_async,
-    /* .get_tensor_2d_async     = */ ggml_backend_cuda_set_tensor_2d_async,
-    /* .set_tensor_2d_async     = */ ggml_backend_cuda_get_tensor_2d_async,
+    /* .set_tensor_2d_async     = */ ggml_backend_cuda_set_tensor_2d_async,
+    /* .get_tensor_2d_async     = */ ggml_backend_cuda_get_tensor_2d_async,
     /* .cpy_tensor_async        = */ ggml_backend_cuda_cpy_tensor_async,
     /* .synchronize             = */ ggml_backend_cuda_synchronize,
     /* .graph_plan_create       = */ NULL,
@@ -5391,8 +5431,8 @@ ggml_backend_reg_t ggml_backend_cuda_reg() {
                 CUDA_CHECK(cudaGetDeviceProperties(&prop, i));
                 dev_ctx->description = prop.name;
 
-                char pci_bus_id[16] = {};
-                snprintf(pci_bus_id, sizeof(pci_bus_id), "%04x:%02x:%02x.0", prop.pciDomainID, prop.pciBusID, prop.pciDeviceID);
+                char pci_bus_id[32] = {};
+                CUDA_CHECK(cudaDeviceGetPCIBusId(pci_bus_id, sizeof(pci_bus_id), i));
                 dev_ctx->pci_bus_id = pci_bus_id;
                 dev_ctx->op_offload_min_batch_size = min_batch_size;
 

ggml/src/ggml-cuda/mma.cuh

@@ -1015,25 +1015,35 @@ namespace ggml_cuda_mma {
 #endif // AMD_MFMA_AVAILABLE
     }
 
-    static __device__ __forceinline__ void mma_block_scaled(tile<16, 8, float> &     D,
-                                                            const tile<16, 8, int> & A,
-                                                            const tile<8, 8, int> &  B,
-                                                            uint32_t                 a_scale,
-                                                            uint32_t                 b_scale) {
+    template <ggml_type type>
+    static __device__ __forceinline__ void mma_block_scaled_fp4(tile<16, 8, float> &     D,
+                                                                const tile<16, 8, int> & A,
+                                                                const tile<8, 8, int> &  B,
+                                                                uint32_t                 a_scale,
+                                                                uint32_t                 b_scale) {
 #ifdef BLACKWELL_MMA_AVAILABLE
         const int * Axi = (const int *) A.x;
         const int * Bxi = (const int *) B.x;
         float *     Dxi = (float *) D.x;
 
-        asm volatile(
-            "mma.sync.aligned.kind::mxf4.block_scale.scale_vec::2X.m16n8k64.row.col.f32.e2m1.e2m1.f32.ue8m0 "
-            "{%0, %1, %2, %3}, {%4, %5, %6, %7}, {%8, %9}, {%0, %1, %2, %3}, "
-            "%10, {0, 0}, %11, {0, 0};"
-            : "+f"(Dxi[0]), "+f"(Dxi[1]), "+f"(Dxi[2]), "+f"(Dxi[3])
-            : "r"(Axi[0]), "r"(Axi[1]), "r"(Axi[2]), "r"(Axi[3]), "r"(Bxi[0]), "r"(Bxi[1]), "r"(a_scale), "r"(b_scale));
+        if constexpr (type == GGML_TYPE_MXFP4) {
+            asm volatile(
+                "mma.sync.aligned.kind::mxf4.block_scale.scale_vec::2X.m16n8k64.row.col.f32.e2m1.e2m1.f32.ue8m0 "
+                "{%0, %1, %2, %3}, {%4, %5, %6, %7}, {%8, %9}, {%0, %1, %2, %3}, "
+                "%10, {0, 0}, %11, {0, 0};"
+                : "+f"(Dxi[0]), "+f"(Dxi[1]), "+f"(Dxi[2]), "+f"(Dxi[3])
+                : "r"(Axi[0]), "r"(Axi[1]), "r"(Axi[2]), "r"(Axi[3]), "r"(Bxi[0]), "r"(Bxi[1]), "r"(a_scale), "r"(b_scale));
+        } else {
+            asm volatile(
+                "mma.sync.aligned.kind::mxf4nvf4.block_scale.scale_vec::4X.m16n8k64.row.col.f32.e2m1.e2m1.f32.ue4m3 "
+                "{%0, %1, %2, %3}, {%4, %5, %6, %7}, {%8, %9}, {%0, %1, %2, %3}, "
+                "%10, {0, 0}, %11, {0, 0};"
+                : "+f"(Dxi[0]), "+f"(Dxi[1]), "+f"(Dxi[2]), "+f"(Dxi[3])
+                : "r"(Axi[0]), "r"(Axi[1]), "r"(Axi[2]), "r"(Axi[3]), "r"(Bxi[0]), "r"(Bxi[1]), "r"(a_scale), "r"(b_scale));
+        }
 #else
         GGML_UNUSED_VARS(D, A, B, a_scale, b_scale);
-#endif  // BLACKWELL_MMA_AVAILABLE
+#endif // BLACKWELL_MMA_AVAILABLE
     }
 
     static __device__ __forceinline__ void mma(

ggml/src/ggml-cuda/mmq.cu

@@ -122,7 +122,7 @@ void ggml_cuda_mul_mat_q(
                             || GGML_CUDA_CC_IS_CDNA(cc);
 
     // TODO: tighter pool buffer size vs q8 path
-    const bool use_native_mxfp4 = blackwell_mma_available(cc) && src0->type == GGML_TYPE_MXFP4;
+    const bool use_native_fp4 = blackwell_mma_available(cc) && (src0->type == GGML_TYPE_MXFP4 || src0->type == GGML_TYPE_NVFP4);
 
     if (!ids) {
         const size_t nbytes_src1_q8_1 = ne13*ne12 * ne11*ne10_padded * sizeof(block_q8_1)/QK8_1 +
@@ -133,9 +133,9 @@ void ggml_cuda_mul_mat_q(
             const int64_t s11 = src1->nb[1] / ts_src1;
             const int64_t s12 = src1->nb[2] / ts_src1;
             const int64_t s13 = src1->nb[3] / ts_src1;
-            if (use_native_mxfp4) {
+            if (use_native_fp4) {
                 static_assert(sizeof(block_fp4_mmq) == 4 * sizeof(block_q8_1));
-                quantize_mmq_mxfp4_cuda(src1_d, nullptr, src1_q8_1.get(), src0->type, ne10, s11, s12, s13, ne10_padded,
+                quantize_mmq_fp4_cuda(src1_d, nullptr, src1_q8_1.get(), src0->type, ne10, s11, s12, s13, ne10_padded,
                                         ne11, ne12, ne13, stream);
 
             } else {
@@ -146,10 +146,8 @@ void ggml_cuda_mul_mat_q(
         }
 
         // Stride depends on quantization format
-        const int64_t s12 = use_native_mxfp4 ?
-                                ne11 * ne10_padded * sizeof(block_fp4_mmq) /
-                                    (8 * QK_MXFP4 * sizeof(int))  // block_fp4_mmq holds 256 values (8 blocks of 32)
-                                :
+        const int64_t s12 = use_native_fp4 ?
+                                ne11 * ne10_padded * sizeof(block_fp4_mmq) / (QK_K * sizeof(int)) :  // block_fp4_mmq holds 256 values
                                 ne11 * ne10_padded * sizeof(block_q8_1) / (QK8_1 * sizeof(int));
         const int64_t s13 = ne12*s12;
 
@@ -198,8 +196,8 @@ void ggml_cuda_mul_mat_q(
         const int64_t s12 = src1->nb[2] / ts_src1;
         const int64_t s13 = src1->nb[3] / ts_src1;
 
-        if (use_native_mxfp4) {
-            quantize_mmq_mxfp4_cuda(src1_d, ids_src1.get(), src1_q8_1.get(), src0->type, ne10, s11, s12, s13,
+        if (use_native_fp4) {
+            quantize_mmq_fp4_cuda(src1_d, ids_src1.get(), src1_q8_1.get(), src0->type, ne10, s11, s12, s13,
                                     ne10_padded, ne11_flat, ne12_flat, ne13_flat, stream);
         } else {
             quantize_mmq_q8_1_cuda(src1_d, ids_src1.get(), src1_q8_1.get(), src0->type, ne10, s11, s12, s13,
@@ -208,8 +206,9 @@ void ggml_cuda_mul_mat_q(
         CUDA_CHECK(cudaGetLastError());
     }
 
-    const int64_t s12 = use_native_mxfp4 ? ne11 * ne10_padded * sizeof(block_fp4_mmq) / (8 * QK_MXFP4 * sizeof(int)) :
-                                           ne11 * ne10_padded * sizeof(block_q8_1) / (QK8_1 * sizeof(int));
+    static_assert(QK_K == 8 * QK_MXFP4, "QK_K needs to be 8 * QK_MXFP4");
+    const int64_t s12 = use_native_fp4 ? ne11 * ne10_padded * sizeof(block_fp4_mmq) / (QK_K * sizeof(int)) :
+                                         ne11 * ne10_padded * sizeof(block_q8_1) / (QK8_1 * sizeof(int));
     const int64_t s13 = ne12*s12;
 
     // Note that ne02 is used instead of ne12 because the number of y channels determines the z dimension of the CUDA grid.

ggml/src/ggml-cuda/mmq.cuh

@@ -10,9 +10,9 @@
 using namespace ggml_cuda_mma;
 
 #define MMQ_DP4A_MAX_BATCH_SIZE 64 // Max. batch size to use for dp4a MMQ kernels when FP16 tensor cores are available.
-#define MMQ_ITER_K 256
-#define MMQ_ITER_K_MXFP4_FP4    512
-#define MMQ_NWARPS 8
+#define MMQ_ITER_K             256
+#define MMQ_ITER_K_FP4         512
+#define MMQ_NWARPS               8
 
 typedef void (*load_tiles_mmq_t)(const char * __restrict__ x, int * x_tile, const int kbx0, const int i_max, const int stride);
 typedef void (*vec_dot_mmq_t)(const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int k00);
@@ -46,9 +46,12 @@ struct block_q8_1_mmq {
     int8_t qs[4*QK8_1]; // 128 values quantized to 8 bit each
 };
 
+// this struct is used for fp4 data types (currently only used for Blackwell)
+// mxfp4 has block size 32, each int32 of d4 contains 2 e8m0 scales in the lower 16 bits
+// nvfp4 has block size 16, each int32 of d4 contains 4 ue4m3 scales
 struct block_fp4_mmq {
-    uint32_t d4[4];       // 8 E8M0 scales (1 per 32 values), 2 packed per uint32: d4[0]={s0,s1}, d4[1]={s2,s3}, etc.
-    int8_t   qs[4 * 32];  // 256 FP4 values packed as 4-bit pairs (2 per byte), 8 blocks of 32 values
+    uint32_t d4[4];
+    int8_t   qs[4 * 32];  // 256 FP4 values packed as 4-bit pairs (2 per byte)
 };
 
 static_assert(sizeof(block_q8_1_mmq) == 4*QK8_1 + 4*sizeof(half2), "Unexpected block_q8_1_mmq size");
@@ -143,10 +146,11 @@ static int get_mmq_y_host(const int cc) {
 
 static constexpr __device__ int get_iter_k([[maybe_unused]] const ggml_type type) {
 #if defined(BLACKWELL_MMA_AVAILABLE)
-    return type == GGML_TYPE_MXFP4 ? MMQ_ITER_K_MXFP4_FP4 : MMQ_ITER_K;
-#else
-    return MMQ_ITER_K;
+if (type == GGML_TYPE_NVFP4 || type == GGML_TYPE_MXFP4) {
+    return MMQ_ITER_K_FP4;
+}
 #endif // defined(BLACKWELL_MMA_AVAILABLE)
+    return MMQ_ITER_K;
 }
 
 static constexpr __device__ int get_mmq_y_device() {
@@ -213,8 +217,8 @@ static constexpr __host__ __device__ tile_x_sizes mmq_get_dp4a_tile_x_sizes(ggml
 }
 
 #define MMQ_MMA_TILE_X_K_Q8_0  (2*MMQ_TILE_NE_K + 2*MMQ_TILE_NE_K/QI8_0                   + 4)
-#define MMQ_MMA_TILE_X_K_FP4   (2*MMQ_TILE_NE_K + 8                                       + 4) // MXFP4
-#define MMQ_MMA_TILE_X_K_NVFP4 (2*MMQ_TILE_NE_K + MMQ_TILE_NE_K/2                         + 4) // NVFP4
+#define MMQ_MMA_TILE_X_K_FP4   (2*MMQ_TILE_NE_K + 8                                       + 4) // MXFP4 and NVFP4 Blackwell
+#define MMQ_MMA_TILE_X_K_NVFP4 (2*MMQ_TILE_NE_K + MMQ_TILE_NE_K/2                         + 4) // NVFP4 Generic
 #define MMQ_MMA_TILE_X_K_Q8_1  (2*MMQ_TILE_NE_K + 2*MMQ_TILE_NE_K/QI8_0                   + 4)
 #define MMQ_MMA_TILE_X_K_Q2_K  (2*MMQ_TILE_NE_K + MMQ_TILE_NE_K                           + 4)
 #define MMQ_MMA_TILE_X_K_Q3_K  (2*MMQ_TILE_NE_K + MMQ_TILE_NE_K/2                         + 4)
@@ -240,7 +244,11 @@ static constexpr __host__ __device__ int mmq_get_mma_tile_x_k(ggml_type type) {
         case GGML_TYPE_Q8_0:    return MMQ_MMA_TILE_X_K_Q8_0;
         // tile sizes are the same for Q8_1 and FP4 for blackwell
         case GGML_TYPE_MXFP4:   return MMQ_MMA_TILE_X_K_Q8_1;
+#if defined(BLACKWELL_MMA_AVAILABLE)
+        case GGML_TYPE_NVFP4:   return MMQ_MMA_TILE_X_K_FP4;
+#else
         case GGML_TYPE_NVFP4:   return MMQ_MMA_TILE_X_K_NVFP4;
+#endif // defined(BLACKWELL_MMA_AVAILABLE)
         case GGML_TYPE_Q2_K:    return MMQ_MMA_TILE_X_K_Q2_K;
         case GGML_TYPE_Q3_K:    return MMQ_MMA_TILE_X_K_Q3_K;
         case GGML_TYPE_Q4_K:    return MMQ_MMA_TILE_X_K_Q8_1;
@@ -934,6 +942,128 @@ static __device__ __forceinline__ void load_tiles_mxfp4_fp4(const char * __restr
     }
 }
 
+#ifdef BLACKWELL_MMA_AVAILABLE
+template <int mmq_y, bool need_check>
+static __device__ __forceinline__ void load_tiles_nvfp4_nvfp4(const char * __restrict__ x,
+                                                            int * __restrict__ x_tile,
+                                                            const int kbx0,
+                                                            const int i_max,
+                                                            const int stride) {
+    constexpr int nwarps = mmq_get_nwarps_device();
+    constexpr int warp_size = ggml_cuda_get_physical_warp_size();
+    constexpr int iter_k = get_iter_k(GGML_TYPE_NVFP4);
+    constexpr int threads_per_row = iter_k / QK_NVFP4; // each thread processes 1 block
+    constexpr int rows_per_warp = warp_size / threads_per_row;
+
+    uint32_t * x_u32 = (uint32_t *) x_tile;
+
+    const int txi = threadIdx.x;
+    const int kbx = txi % threads_per_row;
+    const int row_in_warp = txi / threads_per_row;
+
+    const block_nvfp4 * bxi_base = (const block_nvfp4 *) x + kbx0 + kbx;
+    uint32_t * x_u32_scale = x_u32 + 64 + kbx;
+
+#pragma unroll
+    for (int i0 = 0; i0 < mmq_y; i0 += rows_per_warp * nwarps) {
+        int i = i0 + threadIdx.y * rows_per_warp + row_in_warp;
+
+        if constexpr (need_check) {
+            i = min(i, i_max);
+        }
+
+        const block_nvfp4 * bxi = bxi_base + i * stride;
+        const int row_base = i * MMQ_MMA_TILE_X_K_FP4;
+        const int q_base = row_base + 8 * kbx;
+
+        const uint32_t * src_qs = reinterpret_cast<const uint32_t *>(bxi->qs);
+
+#pragma unroll
+        for (int sub = 0; sub < QK_NVFP4 / QK_NVFP4_SUB; ++sub) {
+            x_u32[q_base + 2 * sub + 0] = src_qs[2 * sub + 0];
+            x_u32[q_base + 2 * sub + 1] = src_qs[2 * sub + 1];
+        }
+
+        x_u32_scale[row_base] = get_int_b4(bxi->d, 0);
+    }
+}
+
+// Shared MMA kernel for MXFP4 and NVFP4 on Blackwell.
+// Both quantizations encode values as e2m1 (FP4) and produce one uint32 scale per
+// m16n8k64 MMA call; only the PTX kind (scale_vec::2X ue8m0 vs scale_vec::4X ue4m3)
+// and the per-type stride constant differ.
+template <int mmq_x, int mmq_y, ggml_type type>
+static __device__ __forceinline__ void vec_dot_fp4_fp4_mma(const int * __restrict__ x,
+                                                           const int * __restrict__ y,
+                                                           float * __restrict__ sum,
+                                                           const int k00) {
+    static_assert(type == GGML_TYPE_MXFP4 || type == GGML_TYPE_NVFP4,
+                  "vec_dot_fp4_fp4_mma: type must be MXFP4 or NVFP4");
+
+    typedef tile<16, 8, int>   tile_A;
+    typedef tile<8, 8, int>    tile_B;
+    typedef tile<16, 8, float> tile_C;
+
+    constexpr int stride        = MMQ_MMA_TILE_X_K_FP4;
+    constexpr int granularity   = mmq_get_granularity_device(mmq_x);
+    constexpr int rows_per_warp = 2 * granularity;
+    constexpr int ntx           = rows_per_warp / tile_C::I;
+    constexpr int nfrags        = MMQ_TILE_NE_K / tile_A::J;
+
+    y += (threadIdx.y % ntx) * (tile_C::J * MMQ_TILE_Y_K);
+
+    const int *      x_qs = (const int *) x;
+    const uint32_t * x_sc = (const uint32_t *) (x_qs + 2 * MMQ_TILE_NE_K);
+    const int *      y_qs = (const int *) y + 4;
+    const uint32_t * y_sc = (const uint32_t *) y;
+
+    // 2 threads per quad supply the packed scale register to the block_scale MMA,
+    // see https://docs.nvidia.com/cuda/parallel-thread-execution/#warp-level-block-scaling
+    const int tidx_A = threadIdx.x / 4 + (threadIdx.x % 2) * 8;
+    const int tidx_B = threadIdx.x / 4;
+    const int i0     = (threadIdx.y / ntx) * rows_per_warp;
+
+    tile_A   A[ntx][nfrags];
+    uint32_t scaleA[ntx][nfrags];
+
+#pragma unroll
+    for (int n = 0; n < ntx; ++n) {
+#pragma unroll
+        for (int frag = 0; frag < nfrags; ++frag) {
+            const int k0 = k00 + frag * tile_A::J;
+            load_ldmatrix(A[n][frag], x_qs + (i0 + n * tile_A::I) * stride + k0, stride);
+            scaleA[n][frag] = x_sc[(i0 + n * tile_A::I + tidx_A) * stride + k0 / tile_A::J];
+        }
+    }
+
+#pragma unroll
+    for (int j0 = 0; j0 < mmq_x; j0 += ntx * tile_C::J) {
+        tile_B   B[nfrags];
+        uint32_t scaleB[nfrags];
+
+#pragma unroll
+        for (int frag = 0; frag < nfrags; ++frag) {
+            const int k0 = frag * tile_B::J;
+            load_generic(B[frag], y_qs + j0 * MMQ_TILE_Y_K + k0, MMQ_TILE_Y_K);
+            scaleB[frag] = y_sc[(j0 + tidx_B) * MMQ_TILE_Y_K + frag];
+        }
+
+#pragma unroll
+        for (int n = 0; n < ntx; ++n) {
+#pragma unroll
+            for (int frag = 0; frag < nfrags; ++frag) {
+                tile_C C = {};
+                mma_block_scaled_fp4<type>(C, A[n][frag], B[frag], scaleA[n][frag], scaleB[frag]);
+#pragma unroll
+                for (int l = 0; l < tile_C::ne; ++l) {
+                    sum[(j0 / tile_C::J + n) * tile_C::ne + l] += C.x[l];
+                }
+            }
+        }
+    }
+}
+#endif // BLACKWELL_MMA_AVAILABLE
+
 
 template <int mmq_y, bool need_check>
 static __device__ __forceinline__ void load_tiles_nvfp4(const char * __restrict__ x,
@@ -1163,77 +1293,6 @@ static __device__ __forceinline__ void vec_dot_q8_0_q8_1_mma(
 #endif // defined(AMD_MFMA_AVAILABLE) || defined(AMD_WMMA_AVAILABLE)
 }
 
-template <int mmq_x, int mmq_y>
-static __device__ __forceinline__ void vec_dot_mxfp4_mxfp4_mma(const int * __restrict__ x,
-                                                               const int * __restrict__ y,
-                                                               float * __restrict__ sum,
-                                                               const int k00) {
-    typedef tile<16, 8, int>   tile_A;
-    typedef tile<8, 8, int>    tile_B;
-    typedef tile<16, 8, float> tile_C;  // Output is float for native scaled MMA
-
-    constexpr int granularity   = mmq_get_granularity_device(mmq_x);
-    constexpr int rows_per_warp = 2 * granularity;
-    constexpr int ntx           = rows_per_warp / tile_C::I;  // Number of x minitiles per warp.
-
-    y += (threadIdx.y % ntx) * (tile_C::J * MMQ_TILE_Y_FP4_K);
-
-    // Match layout from load_tiles_mxfp4_fp4
-    const int *      x_qs = (const int *) x;
-    const uint32_t * x_sc = (const uint32_t *) (x_qs + 2 * MMQ_TILE_NE_K);
-    const int *      y_qs = (const int *) y + 4;
-    const uint32_t * y_sc = (const uint32_t *) y;
-
-    // tile_A has a length of 64 logical values vs. 32 values in block_mxfp4
-    tile_A   A[ntx][MMQ_TILE_NE_K / (2 * QI_MXFP4)];
-    uint32_t scaleA[ntx][MMQ_TILE_NE_K / (2 * QI_MXFP4)];
-
-    // Block scale
-    // Each thread has to point to a 4 byte scale value
-    // https://docs.nvidia.com/cuda/parallel-thread-execution/#warp-level-block-scaling
-
-    const int i0 = (threadIdx.y / ntx) * rows_per_warp;
-
-#pragma unroll
-    for (int n = 0; n < ntx; ++n) {
-#pragma unroll
-        for (int k01 = 0; k01 < MMQ_TILE_NE_K; k01 += 2 * QI_MXFP4) {
-            const int k0 = k00 + k01;
-
-            load_ldmatrix(A[n][k01 / (2 * QI_MXFP4)], x_qs + (i0 + n * tile_A::I) * MMQ_MMA_TILE_X_K_FP4 + k0,
-                          MMQ_MMA_TILE_X_K_FP4);
-
-            // based on block-scaling document, 2 threads in each quad need to supply to the scale value
-            const int tidx         = threadIdx.x / 4 + (threadIdx.x % 2) * 8;
-            scaleA[n][k01 / (2 * QI_MXFP4)] =
-                *(x_sc + (i0 + n * tile_A::I + tidx) * MMQ_MMA_TILE_X_K_FP4 + k0 / (2 * QI_MXFP4));
-        }
-    }
-
-#pragma unroll
-    for (int j0 = 0; j0 < mmq_x; j0 += ntx * tile_C::J) {
-#pragma unroll
-        for (int k01 = 0; k01 < MMQ_TILE_NE_K; k01 += 2 * QI_MXFP4) {
-            tile_B   B;
-            uint32_t scaleB;  // 2xN scales
-
-            load_generic(B, y_qs + j0 * MMQ_TILE_Y_FP4_K + k01, MMQ_TILE_Y_FP4_K);
-
-            scaleB = y_sc[(j0 + threadIdx.x / 4) * MMQ_TILE_Y_FP4_K + k01 / (2 * QI_MXFP4)];
-
-#pragma unroll
-            for (int n = 0; n < ntx; ++n) {
-                tile_C C;
-
-                mma_block_scaled(C, A[n][k01 / (2 * QI_MXFP4)], B, scaleA[n][k01 / (2 * QI_MXFP4)], scaleB);
-#pragma unroll
-                for (int l = 0; l < tile_C::ne; ++l) {
-                    sum[(j0 / tile_C::J + n) * tile_C::ne + l] += C.x[l];
-                }
-            }
-        }
-    }
-}
 
 template <int mmq_x, int mmq_y>
 static __device__ __forceinline__ void vec_dot_q8_1_q8_1_dp4a(
@@ -3259,7 +3318,7 @@ struct mmq_type_traits<mmq_x, mmq_y, need_check, GGML_TYPE_MXFP4> {
     static constexpr int              vdr          = VDR_MXFP4_Q8_1_MMQ;
 #ifdef BLACKWELL_MMA_AVAILABLE
     static constexpr load_tiles_mmq_t load_tiles  = load_tiles_mxfp4_fp4<mmq_y, need_check>;
-    static constexpr vec_dot_mmq_t    vec_dot_mma = vec_dot_mxfp4_mxfp4_mma<mmq_x, mmq_y>;
+    static constexpr vec_dot_mmq_t    vec_dot_mma = vec_dot_fp4_fp4_mma<mmq_x, mmq_y, GGML_TYPE_MXFP4>;
 #else
     static constexpr load_tiles_mmq_t load_tiles   = load_tiles_mxfp4<mmq_y, need_check>;
     static constexpr vec_dot_mmq_t    vec_dot_mma  = vec_dot_q8_0_q8_1_mma<mmq_x, mmq_y, MMQ_Q8_1_DS_LAYOUT_D4>;
@@ -3270,8 +3329,13 @@ struct mmq_type_traits<mmq_x, mmq_y, need_check, GGML_TYPE_MXFP4> {
 template <int mmq_x, int mmq_y, bool need_check>
 struct mmq_type_traits<mmq_x, mmq_y, need_check, GGML_TYPE_NVFP4> {
     static constexpr int              vdr          = VDR_NVFP4_Q8_1_MMQ;
+#ifdef BLACKWELL_MMA_AVAILABLE
+    static constexpr load_tiles_mmq_t load_tiles   = load_tiles_nvfp4_nvfp4<mmq_y, need_check>;
+    static constexpr vec_dot_mmq_t    vec_dot_mma  = vec_dot_fp4_fp4_mma<mmq_x, mmq_y, GGML_TYPE_NVFP4>;
+#else
     static constexpr load_tiles_mmq_t load_tiles   = load_tiles_nvfp4<mmq_y, need_check>;
     static constexpr vec_dot_mmq_t    vec_dot_mma  = vec_dot_q8_0_16_q8_1_mma<mmq_x, mmq_y>;
+#endif // BLACKWELL_MMA_AVAILABLE
     static constexpr vec_dot_mmq_t    vec_dot_dp4a = vec_dot_q8_0_16_q8_1_dp4a<mmq_x, mmq_y>;
 };
 
@@ -3406,7 +3470,7 @@ static __device__ __forceinline__ void mul_mat_q_process_tile(
 
 #if defined(BLACKWELL_MMA_AVAILABLE)
     // FP4 tile stores 8 blocks
-    constexpr int ne_block = (type == GGML_TYPE_MXFP4) ? 8 * QK_MXFP4 : 4 * QK8_1;
+    constexpr int ne_block = (type == GGML_TYPE_MXFP4 || type == GGML_TYPE_NVFP4) ? QK_K : 4 * QK8_1;
 #else
     constexpr int ne_block = 4 * QK8_1;
 #endif  // defined(BLACKWELL_MMA_AVAILABLE)
@@ -3478,10 +3542,10 @@ template <ggml_type type, int mmq_x, bool need_check>
 static __global__ void mul_mat_q(
         const char * __restrict__ x, const int * __restrict__ y, const int32_t * __restrict__ ids_dst,
         const int32_t * __restrict__ expert_bounds, float * __restrict__ dst, float * __restrict__ tmp_fixup,
-        const int ncols_x, const int nrows_x, const int ncols_dst, const int stride_row_x, const int ncols_y, const int stride_col_dst,
-        const int channel_ratio, const int nchannels_y, const int stride_channel_x, const int stride_channel_y, const int stride_channel_dst,
-        const int sample_ratio, const int nsamples_y, const int stride_sample_x, const int stride_sample_y, const int stride_sample_dst,
-        const int ncols_max) {
+        const uint3 blocks_per_ne00, const int nrows_x, const int ncols_dst, const int stride_row_x, const int ncols_y, const int stride_col_dst,
+        const uint3 channel_ratio, const uint3 nchannels_y, const int stride_channel_x, const int stride_channel_y, const int stride_channel_dst,
+        const uint3 sample_ratio, const uint3 nsamples_y, const int stride_sample_x, const int stride_sample_y, const int stride_sample_dst,
+        const uint3 ntx) {
 
     // Skip unused template specializations for faster compilation:
     if (mmq_x > get_mmq_x_max_device() || mmq_x % mmq_get_granularity_device(mmq_x) != 0) {
@@ -3495,8 +3559,7 @@ static __global__ void mul_mat_q(
     constexpr int qk    = ggml_cuda_type_traits<type>::qk;
     constexpr int mmq_y = get_mmq_y_device();
 
-    const int ntx = (ncols_max + mmq_x - 1) / mmq_x; // Number of tiles x
-    const int nty = (nrows_x   + mmq_y - 1) / mmq_y; // Number of tiles y
+    const uint32_t nty = (nrows_x + mmq_y - 1) / mmq_y; // Number of tiles y
 
     // Initialize the ids for writing back data with just the index.
     // For regular matrix multiplications this is never changed.
@@ -3517,8 +3580,9 @@ static __global__ void mul_mat_q(
     // On non-CDNA AMD or old CUDA the performance with stream-k was worse, use conventional tiling instead:
 #if (defined(GGML_USE_HIP) && !defined(CDNA)) || __CUDA_ARCH__ < GGML_CUDA_CC_VOLTA
     {
-        const int wt = blockIdx.z / nchannels_y;
-        const int zt = blockIdx.z - wt*nchannels_y;
+        const uint2 tmp2 = fast_div_modulo(blockIdx.z, nchannels_y);
+        const int wt = tmp2.x;
+        const int zt = tmp2.y;
         const int jt = blockIdx.y;
         const int it = blockIdx.x;
 
@@ -3561,40 +3625,40 @@ static __global__ void mul_mat_q(
         const int tile_x_max_i = nrows_x  - it*mmq_y - 1;
         const int tile_y_max_j = col_diff - jt*mmq_x - 1;
 
-        const int offset_x = (wt/sample_ratio)*stride_sample_x + (zt/channel_ratio)*stride_channel_x + it*mmq_y*stride_row_x;
+        const int offset_x = fastdiv(wt, sample_ratio)*stride_sample_x + fastdiv(zt, channel_ratio)*stride_channel_x + it*mmq_y*stride_row_x;
 
         constexpr bool fixup = false;
         mul_mat_q_process_tile<type, mmq_x, need_check, fixup>
             (x, offset_x, y + offset_y, ids_dst_shared, dst + offset_dst, tmp_fixup, stride_row_x, ncols_y, stride_col_dst,
-             tile_x_max_i, tile_y_max_j, 0, ncols_x/qk);
+             tile_x_max_i, tile_y_max_j, 0, blocks_per_ne00.z);
         return;
     }
 #endif // (defined(GGML_USE_HIP) && !defined(CDNA4) && !defined(CDNA3)) || __CUDA_ARCH__ < GGML_CUDA_CC_VOLTA
 
-    constexpr int ITER_K = get_iter_k(type);
-
-    const     int64_t blocks_per_ne00 = ncols_x / qk;
-    constexpr int     blocks_per_iter = ITER_K / qk;
+    constexpr int ITER_K          = get_iter_k(type);
+    constexpr int blocks_per_iter = ITER_K / qk;
 
     // kbc == k block continuous, current index in continuous ijk space.
-    int64_t kbc      = (int64_t) blockIdx.x     *nsamples_y*nchannels_y*ntx*nty*blocks_per_ne00 / gridDim.x;
-    int64_t kbc_stop = (int64_t)(blockIdx.x + 1)*nsamples_y*nchannels_y*ntx*nty*blocks_per_ne00 / gridDim.x;
+    int kbc      = int64_t(blockIdx.x)    *(nsamples_y.z*nchannels_y.z*ntx.z*nty*blocks_per_ne00.z) / gridDim.x;
+    int kbc_stop = int64_t(blockIdx.x + 1)*(nsamples_y.z*nchannels_y.z*ntx.z*nty*blocks_per_ne00.z) / gridDim.x;
 
-    kbc      -= (kbc      % blocks_per_ne00) % blocks_per_iter;
-    kbc_stop -= (kbc_stop % blocks_per_ne00) % blocks_per_iter;
+    kbc      -= fastmodulo(kbc,      blocks_per_ne00) % blocks_per_iter;
+    kbc_stop -= fastmodulo(kbc_stop, blocks_per_ne00) % blocks_per_iter;
 
     // kb0 == k index when doing the matrix multiplication for an output tile.
-    int kb0_start = kbc % blocks_per_ne00;
-    int kb0_stop  = min(blocks_per_ne00, kb0_start + kbc_stop - kbc);
-    while (kbc < kbc_stop && kb0_stop == blocks_per_ne00) {
-        int tmp = kbc;
-        const int it = tmp / (nsamples_y*nchannels_y*ntx*blocks_per_ne00);
-        tmp -= it * (nsamples_y*nchannels_y*ntx*blocks_per_ne00);
-        const int wt = tmp / (nchannels_y*ntx*blocks_per_ne00);
-        tmp -= wt * (nchannels_y*ntx*blocks_per_ne00);
-        const int zt = tmp / (ntx*blocks_per_ne00);
-        tmp -= zt * (ntx*blocks_per_ne00);
-        const int jt = tmp / blocks_per_ne00;
+    int kb0_start = fastmodulo(kbc, blocks_per_ne00);
+    int kb0_stop  = min(blocks_per_ne00.z, uint32_t(kb0_start + kbc_stop - kbc));
+    while (kbc < kbc_stop && kb0_stop == int(blocks_per_ne00.z)) {
+        int tmp = fastdiv(kbc, blocks_per_ne00);
+        uint2 tmp2 = fast_div_modulo(tmp, ntx);
+        const int jt = tmp2.y;
+        tmp = tmp2.x;
+        tmp2 = fast_div_modulo(tmp, nchannels_y);
+        const int zt = tmp2.y;
+        tmp = tmp2.x;
+        tmp2 = fast_div_modulo(tmp, nsamples_y);
+        const int wt = tmp2.y;
+        const int it = tmp2.x;
 
         // Defaults for regular matrix multiplication:
         int col_low    = 0;
@@ -3612,11 +3676,11 @@ static __global__ void mul_mat_q(
             offset_dst = 0;
 
             if (jt*mmq_x >= col_diff) {
-                kbc += blocks_per_ne00;
-                kbc -= kbc % blocks_per_ne00;
+                kbc += blocks_per_ne00.z;
+                kbc -= fastmodulo(kbc, blocks_per_ne00);
 
                 kb0_start = 0;
-                kb0_stop  = min(blocks_per_ne00, kbc_stop - kbc);
+                kb0_stop  = min(blocks_per_ne00.z, uint32_t(kbc_stop - kbc));
 
                 continue;
             }
@@ -3641,32 +3705,34 @@ static __global__ void mul_mat_q(
         const int tile_x_max_i = nrows_x  - it*mmq_y - 1;
         const int tile_y_max_j = col_diff - jt*mmq_x - 1;
 
-        const int offset_x = (wt/sample_ratio)*stride_sample_x + (zt/channel_ratio)*stride_channel_x + it*mmq_y*stride_row_x;
+        const int offset_x = fastdiv(wt, sample_ratio)*stride_sample_x + fastdiv(zt, channel_ratio)*stride_channel_x + it*mmq_y*stride_row_x;
 
         constexpr bool fixup = false; // All but (potentially) the last iterations write their data to dst rather than the fixup buffer.
         mul_mat_q_process_tile<type, mmq_x, need_check, fixup>
             (x, offset_x, y + offset_y, ids_dst_shared, dst + offset_dst, tmp_fixup, stride_row_x, ncols_y, stride_col_dst,
              tile_x_max_i, tile_y_max_j, kb0_start, kb0_stop);
 
-        kbc += blocks_per_ne00;
-        kbc -= kbc % blocks_per_ne00;
+        kbc += blocks_per_ne00.z;
+        kbc -= fastmodulo(kbc, blocks_per_ne00);
 
         kb0_start = 0;
-        kb0_stop  = min(blocks_per_ne00, kbc_stop - kbc);
+        kb0_stop  = min(blocks_per_ne00.z, uint32_t(kbc_stop - kbc));
     }
 
     if (kbc >= kbc_stop) {
         return;
     }
 
-    int tmp = kbc;
-    const int it = tmp / (nsamples_y*nchannels_y*ntx*blocks_per_ne00);
-    tmp -= it * (nsamples_y*nchannels_y*ntx*blocks_per_ne00);
-    const int wt = tmp / (nchannels_y*ntx*blocks_per_ne00);
-    tmp -= wt * (nchannels_y*ntx*blocks_per_ne00);
-    const int zt = tmp / (ntx*blocks_per_ne00);
-    tmp -= zt * (ntx*blocks_per_ne00);
-    const int jt = tmp / blocks_per_ne00;
+    int tmp = fastdiv(kbc, blocks_per_ne00);
+    uint2 tmp2 = fast_div_modulo(tmp, ntx);
+    const int jt = tmp2.y;
+    tmp = tmp2.x;
+    tmp2 = fast_div_modulo(tmp, nchannels_y);
+    const int zt = tmp2.y;
+    tmp = tmp2.x;
+    tmp2 = fast_div_modulo(tmp, nsamples_y);
+    const int wt = tmp2.y;
+    const int it = tmp2.x;
 
     // Defaults for regular matrix multiplication:
     int col_low    = 0;
@@ -3708,7 +3774,7 @@ static __global__ void mul_mat_q(
     const int tile_x_max_i = nrows_x  - it*mmq_y - 1;
     const int tile_y_max_j = col_diff - jt*mmq_x - 1;
 
-    const int offset_x = (wt/sample_ratio)*stride_sample_x + (zt/channel_ratio)*stride_channel_x + it*mmq_y*stride_row_x;
+    const int offset_x = fastdiv(wt, sample_ratio)*stride_sample_x + fastdiv(zt, channel_ratio)*stride_channel_x + it*mmq_y*stride_row_x;
 
     constexpr bool fixup = true; // Last index writes its data to fixup buffer to avoid data races with other blocks.
     mul_mat_q_process_tile<type, mmq_x, need_check, fixup>
@@ -3717,46 +3783,37 @@ static __global__ void mul_mat_q(
 }
 
 template <ggml_type type, int mmq_x, bool need_check>
-static __global__ void mul_mat_q_stream_k_fixup(const int32_t * ids_dst,
-                                                const int32_t * expert_bounds,
-                                                float * __restrict__ dst,
-                                                const float * __restrict__ tmp_last_tile,
-                                                const int    ncols_x,
-                                                const int    nrows_x,
-                                                const int    ncols_dst,
-                                                const size_t stride_col_dst,
-                                                const int    nchannels_y,
-                                                const size_t stride_channel_dst,
-                                                const int    nsamples_y,
-                                                const size_t stride_sample_dst,
-                                                const int    ncols_max) {
-    constexpr int     mmq_y           = get_mmq_y_device();
-    constexpr int     qk              = ggml_cuda_type_traits<type>::qk;
-    constexpr int     ITER_K          = get_iter_k(type);
+__launch_bounds__(ggml_cuda_get_physical_warp_size()*mmq_get_nwarps_device()/2, 1)
+static __global__ void mul_mat_q_stream_k_fixup(
+        const int32_t * __restrict__ ids_dst, const int32_t * __restrict__ expert_bounds, float * __restrict__ dst,
+        float * __restrict__ tmp_last_tile, const uint3 blocks_per_ne00, const int nrows_x, const int ncols_dst,
+        const int stride_col_dst, const uint3 nchannels_y, const int stride_channel_dst, const uint3 nsamples_y,
+        const int stride_sample_dst, const uint3 ntx) {
+    constexpr int mmq_y           = get_mmq_y_device();
+    constexpr int qk              = ggml_cuda_type_traits<type>::qk;
+    constexpr int ITER_K          = get_iter_k(type);
+    constexpr int blocks_per_iter = ITER_K / qk;
 
-    constexpr int     blocks_per_iter = ITER_K / qk;
-    const     int64_t blocks_per_ne00 = ncols_x / qk;
-
-    constexpr int nwarps = mmq_get_nwarps_device();
+    constexpr int nwarps = mmq_get_nwarps_device()/2;
     constexpr int warp_size = ggml_cuda_get_physical_warp_size();
 
-    float sum[mmq_x*mmq_y / (nwarps*warp_size)] = {0.0f};
+    float sum[mmq_x / nwarps] = {0.0f};
+    const int i = blockIdx.y*warp_size + threadIdx.x;
 
-    const int ntx  = (ncols_max + mmq_x - 1) / mmq_x;
-    const int nty  = (nrows_x   + mmq_y - 1) / mmq_y;
+    const int nty = (nrows_x + mmq_y - 1) / mmq_y;
 
     const int bidx0 = blockIdx.x;
 
     // kbc == k block continuous, current index in continuous ijk space.
-    int64_t kbc0      = (int64_t) bidx0     *nsamples_y*nchannels_y*ntx*nty*blocks_per_ne00 / gridDim.x;
-    int64_t kbc0_stop = (int64_t)(bidx0 + 1)*nsamples_y*nchannels_y*ntx*nty*blocks_per_ne00 / gridDim.x;
+    int kbc0      = int64_t(blockIdx.x)    *(nsamples_y.z*nchannels_y.z*ntx.z*nty*blocks_per_ne00.z) / gridDim.x;
+    int kbc0_stop = int64_t(blockIdx.x + 1)*(nsamples_y.z*nchannels_y.z*ntx.z*nty*blocks_per_ne00.z) / gridDim.x;
 
-    kbc0      -= (kbc0      % blocks_per_ne00) % blocks_per_iter;
-    kbc0_stop -= (kbc0_stop % blocks_per_ne00) % blocks_per_iter;
+    kbc0      -= fastmodulo(kbc0,      blocks_per_ne00) % blocks_per_iter;
+    kbc0_stop -= fastmodulo(kbc0_stop, blocks_per_ne00) % blocks_per_iter;
 
     const bool did_not_have_any_data   = kbc0 == kbc0_stop;
-    const bool wrote_beginning_of_tile = kbc0 % blocks_per_ne00 == 0;
-    const bool did_not_write_last      = kbc0/blocks_per_ne00 == kbc0_stop/blocks_per_ne00 && kbc0_stop % blocks_per_ne00 != 0;
+    const bool wrote_beginning_of_tile = fastmodulo(kbc0, blocks_per_ne00) == 0;
+    const bool did_not_write_last      = fastdiv(kbc0, blocks_per_ne00) == fastdiv(kbc0_stop, blocks_per_ne00) && fastmodulo(kbc0_stop, blocks_per_ne00) != 0;
     if (did_not_have_any_data || wrote_beginning_of_tile || did_not_write_last) {
         return;
     }
@@ -3765,11 +3822,11 @@ static __global__ void mul_mat_q_stream_k_fixup(const int32_t * ids_dst,
 
     // Iterate over previous blocks and sum up partial sums written to fixup buffer.
     // All CUDA blocks that get here must have a previous block that needs a fixup.
-    int64_t bidx = bidx0 - 1;
-    int64_t kbc_stop = kbc0;
+    int bidx = bidx0 - 1;
+    int kbc_stop = kbc0;
     while(true) {
-        int64_t kbc = bidx*nsamples_y*nchannels_y*ntx*nty*blocks_per_ne00 / gridDim.x;
-        kbc -= (kbc % blocks_per_ne00) % blocks_per_iter;
+        int kbc = int64_t(bidx)*(nsamples_y.z*nchannels_y.z*ntx.z*nty*blocks_per_ne00.z) / gridDim.x;
+        kbc -= fastmodulo(kbc, blocks_per_ne00) % blocks_per_iter;
 
         if (kbc == kbc_stop) { // Did not have any data.
             bidx--;
@@ -3779,20 +3836,16 @@ static __global__ void mul_mat_q_stream_k_fixup(const int32_t * ids_dst,
 
         any_fixup = true;
 
+
 #pragma unroll
         for (int j0 = 0; j0 < mmq_x; j0 += nwarps) {
             const int j = j0 + threadIdx.y;
 
-#pragma unroll
-            for (int i0 = 0; i0 < mmq_y; i0 += warp_size) {
-                const int i = i0 + threadIdx.x;
-
-                sum[(j0/nwarps) * (mmq_y/warp_size) + i0/warp_size] += tmp_last_tile[bidx*(mmq_x*mmq_y) + j*mmq_y + i];
-            }
+            sum[j0/nwarps] += tmp_last_tile[bidx*(mmq_x*mmq_y) + j*mmq_y + i];
         }
 
         // If this block started in a previous tile we are done and don't need to combine additional partial results.
-        if (kbc % blocks_per_ne00 == 0 || kbc/blocks_per_ne00 < kbc0/blocks_per_ne00) {
+        if (fastmodulo(kbc, blocks_per_ne00) == 0 || fastdiv(kbc, blocks_per_ne00) < fastdiv(kbc0, blocks_per_ne00)) {
             break;
         }
         bidx--;
@@ -3803,14 +3856,16 @@ static __global__ void mul_mat_q_stream_k_fixup(const int32_t * ids_dst,
         return;
     }
 
-    int tmp = kbc0;
-    const int it = tmp / (nsamples_y*nchannels_y*ntx*blocks_per_ne00);
-    tmp -= it * (nsamples_y*nchannels_y*ntx*blocks_per_ne00);
-    const int wt = tmp / (nchannels_y*ntx*blocks_per_ne00);
-    tmp -= wt * (nchannels_y*ntx*blocks_per_ne00);
-    const int zt = tmp / (ntx*blocks_per_ne00);
-    tmp -= zt * (ntx*blocks_per_ne00);
-    const int jt = tmp / blocks_per_ne00;
+    int tmp = fastdiv(kbc0, blocks_per_ne00);
+    uint2 tmp2 = fast_div_modulo(tmp, ntx);
+    const int jt = tmp2.y;
+    tmp = tmp2.x;
+    tmp2 = fast_div_modulo(tmp, nchannels_y);
+    const int zt = tmp2.y;
+    tmp = tmp2.x;
+    tmp2 = fast_div_modulo(tmp, nsamples_y);
+    const int wt = tmp2.y;
+    const int it = tmp2.x;
 
     if (!ids_dst) {
         const int offset_dst = wt*stride_sample_dst + zt*stride_channel_dst + jt*mmq_x*stride_col_dst + it*mmq_y;
@@ -3818,6 +3873,9 @@ static __global__ void mul_mat_q_stream_k_fixup(const int32_t * ids_dst,
 
         const int i_max = nrows_x   - it*mmq_y - 1;
         const int j_max = ncols_dst - jt*mmq_x - 1;
+        if (need_check && i > i_max) {
+            return;
+        }
 
 #pragma unroll
         for (int j0 = 0; j0 < mmq_x; j0 += nwarps) {
@@ -3827,16 +3885,7 @@ static __global__ void mul_mat_q_stream_k_fixup(const int32_t * ids_dst,
                 return;
             }
 
-#pragma unroll
-            for (int i0 = 0; i0 < mmq_y; i0 += warp_size) {
-                const int i = i0 + threadIdx.x;
-
-                if (need_check && i > i_max) {
-                    continue;
-                }
-
-                dst[j*stride_col_dst + i] += sum[(j0/nwarps) * (mmq_y/warp_size) + i0/warp_size];
-            }
+            dst[j*stride_col_dst + i] += sum[j0/nwarps];
         }
         return;
     }
@@ -3856,6 +3905,9 @@ static __global__ void mul_mat_q_stream_k_fixup(const int32_t * ids_dst,
 
     const int i_max = nrows_x  - it*mmq_y - 1;
     const int j_max = col_diff - jt*mmq_x - 1;
+    if (need_check && i > i_max) {
+        return;
+    }
 
 #pragma unroll
     for (int j0 = 0; j0 < mmq_x; j0 += nwarps) {
@@ -3865,16 +3917,7 @@ static __global__ void mul_mat_q_stream_k_fixup(const int32_t * ids_dst,
             return;
         }
 
-#pragma unroll
-        for (int i0 = 0; i0 < mmq_y; i0 += warp_size) {
-            const int i = i0 + threadIdx.x;
-
-            if (need_check && i > i_max) {
-                continue;
-            }
-
-            dst[ids_dst_shared[j]*stride_col_dst + i] += sum[(j0/nwarps) * (mmq_y/warp_size) + i0/warp_size];
-        }
+        dst[ids_dst_shared[j]*stride_col_dst + i] += sum[j0/nwarps];
     }
 }
 
@@ -3922,29 +3965,44 @@ static void launch_mul_mat_q(ggml_backend_cuda_context & ctx, const mmq_args & a
     const int channel_ratio = args.nchannels_y / args.nchannels_x;
     const int sample_ratio  = args.nsamples_y  / args.nsamples_x;
 
+    const uint3 blocks_per_ne00_fd = init_fastdiv_values(args.ncols_x / ggml_cuda_type_traits<type>::qk);
+    const uint3 ntx_fd             = init_fastdiv_values(ntx);
+    const uint3 nchannels_y_fd     = init_fastdiv_values(args.nchannels_y);
+    const uint3 nsamples_y_fd      = init_fastdiv_values(args.nsamples_y);
+    const uint3 channel_ratio_fd   = init_fastdiv_values(channel_ratio);
+    const uint3 sample_ratio_fd    = init_fastdiv_values(sample_ratio);
+
     if (!args.use_stream_k) {
         if (args.nrows_x % mmq_y == 0) {
             constexpr bool need_check = false;
             mul_mat_q<type, mmq_x, need_check><<<block_nums_xy_tiling, block_dims, nbytes_shared, stream>>>
                 (args.x, args.y, args.ids_dst, args.expert_bounds, args.dst, nullptr,
-                 args.ncols_x, args.nrows_x, args.ncols_dst, args.stride_row_x, args.ncols_y, args.nrows_dst,
-                 channel_ratio, args.nchannels_y, args.stride_channel_x, args.stride_channel_y, args.stride_channel_dst,
-                 sample_ratio, args.nsamples_y, args.stride_sample_x, args.stride_sample_y, args.stride_sample_dst,
-                 args.ncols_max);
+                 blocks_per_ne00_fd, args.nrows_x, args.ncols_dst, args.stride_row_x, args.ncols_y, args.nrows_dst,
+                 channel_ratio_fd, nchannels_y_fd, args.stride_channel_x, args.stride_channel_y, args.stride_channel_dst,
+                 sample_ratio_fd, nsamples_y_fd, args.stride_sample_x, args.stride_sample_y, args.stride_sample_dst,
+                 ntx_fd);
         } else {
             constexpr bool need_check = true;
             mul_mat_q<type, mmq_x, need_check><<<block_nums_xy_tiling, block_dims, nbytes_shared, stream>>>
                 (args.x, args.y, args.ids_dst, args.expert_bounds, args.dst, nullptr,
-                 args.ncols_x, args.nrows_x, args.ncols_dst, args.stride_row_x, args.ncols_y, args.nrows_dst,
-                 channel_ratio, args.nchannels_y, args.stride_channel_x, args.stride_channel_y, args.stride_channel_dst,
-                 sample_ratio, args.nsamples_y, args.stride_sample_x, args.stride_sample_y, args.stride_sample_dst,
-                 args.ncols_max);
+                 blocks_per_ne00_fd, args.nrows_x, args.ncols_dst, args.stride_row_x, args.ncols_y, args.nrows_dst,
+                 channel_ratio_fd, nchannels_y_fd, args.stride_channel_x, args.stride_channel_y, args.stride_channel_dst,
+                 sample_ratio_fd, nsamples_y_fd, args.stride_sample_x, args.stride_sample_y, args.stride_sample_dst,
+                 ntx_fd);
         }
         return;
     }
 
-    const dim3 block_nums_stream_k(nsm, 1, 1);
-    const bool fixup_needed = ntx*nty*ntzw % nsm != 0;
+    // For the stream-k kernel it is possible to run it with tiling by setting the number of CUDA blocks equal to the number of tiles.
+    // This is worthwhile if the efficiency of tiling is high and skipping the fixup kernel is more important.
+    const int ntiles_dst = ntx * nty * ntzw;
+    const int tiles_nwaves = (ntiles_dst + nsm - 1) / nsm;
+    const int tiles_efficiency_percent = 100 * ntiles_dst / (nsm*tiles_nwaves);
+    const dim3 block_nums_stream_k(GGML_CUDA_CC_IS_NVIDIA(cc) && tiles_efficiency_percent >= 90 ? ntiles_dst : nsm, 1, 1);
+
+    GGML_ASSERT(ntiles_dst * blocks_per_ne00_fd.z < (1 << 30)); // Assert that variable kbc will not overflow.
+
+    const bool fixup_needed = ntiles_dst % block_nums_stream_k.x != 0;
 
     ggml_cuda_pool & pool = ctx.pool(id);
     ggml_cuda_pool_alloc<float> tmp_fixup(pool);
@@ -3952,40 +4010,45 @@ static void launch_mul_mat_q(ggml_backend_cuda_context & ctx, const mmq_args & a
         tmp_fixup.alloc(block_nums_stream_k.x * mmq_x*mmq_y);
     }
 
+    const dim3 block_nums_fixup(block_nums_stream_k.x, mmq_y/warp_size, 1);
+    const dim3 block_dims_fixup(block_dims.x, block_dims.y/2, block_dims.z);
+
     if (args.nrows_x % mmq_y == 0) {
         constexpr bool need_check = false;
         mul_mat_q<type, mmq_x, need_check><<<block_nums_stream_k, block_dims, nbytes_shared, stream>>>
             (args.x, args.y, args.ids_dst, args.expert_bounds, args.dst, tmp_fixup.ptr,
-             args.ncols_x, args.nrows_x, args.ncols_dst, args.stride_row_x, args.ncols_y, args.nrows_dst,
-             channel_ratio, args.nchannels_y, args.stride_channel_x, args.stride_channel_y, args.stride_channel_dst,
-             sample_ratio, args.nsamples_y, args.stride_sample_x, args.stride_sample_y, args.stride_sample_dst,
-             args.ncols_max);
+             blocks_per_ne00_fd, args.nrows_x, args.ncols_dst, args.stride_row_x, args.ncols_y, args.nrows_dst,
+             channel_ratio_fd, nchannels_y_fd, args.stride_channel_x, args.stride_channel_y, args.stride_channel_dst,
+             sample_ratio_fd, nsamples_y_fd, args.stride_sample_x, args.stride_sample_y, args.stride_sample_dst,
+             ntx_fd);
 
         if (!fixup_needed) {
             return;
         }
 
-        mul_mat_q_stream_k_fixup<type, mmq_x, need_check><<<block_nums_stream_k, block_dims, 0, stream>>>
-            (args.ids_dst, args.expert_bounds, args.dst, tmp_fixup.ptr, args.ncols_x, args.nrows_x, args.ncols_dst,
-             args.nrows_dst, args.nchannels_y, args.stride_channel_dst, args.nsamples_y, args.stride_sample_dst,
-             args.ncols_max);
+        CUDA_CHECK(cudaGetLastError());
+        mul_mat_q_stream_k_fixup<type, mmq_x, need_check><<<block_nums_fixup, block_dims_fixup, 0, stream>>>
+            (args.ids_dst, args.expert_bounds, args.dst, tmp_fixup.ptr, blocks_per_ne00_fd, args.nrows_x, args.ncols_dst,
+             args.nrows_dst, nchannels_y_fd, args.stride_channel_dst, nsamples_y_fd, args.stride_sample_dst,
+             ntx_fd);
     } else {
         constexpr bool need_check = true;
         mul_mat_q<type, mmq_x, need_check><<<block_nums_stream_k, block_dims, nbytes_shared, stream>>>
             (args.x, args.y, args.ids_dst, args.expert_bounds, args.dst, tmp_fixup.ptr,
-             args.ncols_x, args.nrows_x, args.ncols_dst, args.stride_row_x, args.ncols_y, args.nrows_dst,
-             channel_ratio, args.nchannels_y, args.stride_channel_x, args.stride_channel_y, args.stride_channel_dst,
-             sample_ratio, args.nsamples_y, args.stride_sample_x, args.stride_sample_y, args.stride_sample_dst,
-             args.ncols_max);
+             blocks_per_ne00_fd, args.nrows_x, args.ncols_dst, args.stride_row_x, args.ncols_y, args.nrows_dst,
+             channel_ratio_fd, nchannels_y_fd, args.stride_channel_x, args.stride_channel_y, args.stride_channel_dst,
+             sample_ratio_fd, nsamples_y_fd, args.stride_sample_x, args.stride_sample_y, args.stride_sample_dst,
+             ntx_fd);
 
         if (!fixup_needed) {
             return;
         }
 
-        mul_mat_q_stream_k_fixup<type, mmq_x, need_check><<<block_nums_stream_k, block_dims, 0, stream>>>
-            (args.ids_dst, args.expert_bounds, args.dst, tmp_fixup.ptr, args.ncols_x, args.nrows_x, args.ncols_dst,
-             args.nrows_dst, args.nchannels_y, args.stride_channel_dst, args.nsamples_y, args.stride_sample_dst,
-             args.ncols_max);
+        CUDA_CHECK(cudaGetLastError());
+        mul_mat_q_stream_k_fixup<type, mmq_x, need_check><<<block_nums_fixup, block_dims_fixup, 0, stream>>>
+            (args.ids_dst, args.expert_bounds, args.dst, tmp_fixup.ptr, blocks_per_ne00_fd, args.nrows_x, args.ncols_dst,
+             args.nrows_dst, nchannels_y_fd, args.stride_channel_dst, nsamples_y_fd, args.stride_sample_dst,
+             ntx_fd);
     }
 }
 

ggml/src/ggml-cuda/mmvq.cu

@@ -115,6 +115,7 @@ static constexpr __host__ __device__ int get_mmvq_mmid_max_batch_pascal_older(gg
         case GGML_TYPE_IQ4_NL:  return 6;
         case GGML_TYPE_IQ4_XS:  return 5;
         case GGML_TYPE_MXFP4:   return 4;
+        case GGML_TYPE_NVFP4:   return 4;
         case GGML_TYPE_Q2_K:    return 4;
         case GGML_TYPE_Q3_K:    return 4;
         case GGML_TYPE_Q4_0:    return 6;
@@ -135,6 +136,7 @@ static constexpr __host__ __device__ int get_mmvq_mmid_max_batch_turing_plus(ggm
         case GGML_TYPE_IQ3_S:   return 6;
         case GGML_TYPE_IQ3_XXS: return 7;
         case GGML_TYPE_MXFP4:   return 7;
+        case GGML_TYPE_NVFP4:   return 8;
         case GGML_TYPE_Q2_K:    return 7;
         case GGML_TYPE_Q3_K:    return 5;
         default:                return MMVQ_MAX_BATCH_SIZE;
@@ -221,6 +223,7 @@ static constexpr __host__ __device__ int get_mmvq_mmid_max_batch_rdna4(ggml_type
         case GGML_TYPE_IQ4_NL:  return 7;
         case GGML_TYPE_IQ4_XS:  return 5;
         case GGML_TYPE_MXFP4:   return 5;
+        case GGML_TYPE_NVFP4:   return 5;
         case GGML_TYPE_Q3_K:    return 4;
         case GGML_TYPE_Q4_0:    return 7;
         case GGML_TYPE_Q4_1:    return 7;

ggml/src/ggml-cuda/quantize.cu

@@ -70,6 +70,102 @@ __device__ __forceinline__ uint8_t compute_e8m0_scale(float amax) {
     return static_cast<uint8_t>(biased);
 }
 
+
+static __global__ void quantize_mmq_nvfp4(
+        const float * __restrict__ x, const int32_t * __restrict__ ids, void * __restrict__ vy,
+        const int64_t ne00, const int64_t s01, const int64_t s02, const int64_t s03,
+        const int64_t ne0, const int64_t ne1, const int64_t ne2) {
+#if defined(BLACKWELL_MMA_AVAILABLE)
+
+    const int64_t i0_base = ((int64_t) blockDim.x * blockIdx.y + threadIdx.x) * QK_NVFP4_SUB;
+    if (i0_base >= ne0) {
+        return;
+    }
+
+    const int64_t i1 = blockIdx.x;
+    const int64_t i2 = blockIdx.z % ne2;
+    const int64_t i3 = blockIdx.z / ne2;
+    const int64_t i01 = ids ? ids[i1] : i1;
+    const int64_t k_block = i0_base / QK_K;
+    const int64_t blocks_per_col = (ne0 + QK_K - 1) / QK_K;
+    if (k_block >= blocks_per_col) {
+        return;
+    }
+
+    const int64_t ib = blockIdx.z * ((int64_t) blocks_per_col * ne1) + k_block * ne1 + blockIdx.x;
+    block_fp4_mmq * y = (block_fp4_mmq *) vy;
+    block_fp4_mmq * yb = y + ib;
+
+    const int sub = (i0_base % QK_K) / QK_NVFP4_SUB;
+
+    float vals_raw[QK_NVFP4_SUB];
+    float amax_raw = 0.0f;
+    const int64_t base_idx = i3 * s03 + i2 * s02 + i01 * s01;
+#pragma unroll
+    for (int k = 0; k < QK_NVFP4_SUB; k++) {
+        const int64_t i00 = i0_base + k;
+        if (i00 < ne00) {
+            const float v = x[base_idx + i00];
+            vals_raw[k] = v;
+            amax_raw = fmaxf(amax_raw, fabsf(v));
+        } else {
+            vals_raw[k] = 0.0f;
+        }
+    }
+
+    static constexpr int test_offsets[5] = { 0, -1, 1, -2, 2};
+    const int first_fp8_code = (int) ggml_cuda_fp32_to_ue4m3(amax_raw / 6.0f);
+
+    float best_err = FLT_MAX;
+    uint8_t fp8_code = 0;
+    float subblock_scale = 0.0f;
+
+#pragma unroll // Check +/- 2 to find best code to reduce NVFP4 activation loss. Negligible overhead on Blackwell.
+    for (int i = 0; i < 5; i++) {
+        const int test_code = first_fp8_code + test_offsets[i];
+        if (test_code < 0 || test_code > 0x7e) {
+            continue;
+        }
+        const uint8_t code = (uint8_t) test_code;
+        const float test_scale = ggml_cuda_ue4m3_to_fp32(code);
+        const float test_inv_scale = test_scale > 0.0f ? 0.5f / test_scale : 0.0f;
+        float cur_err = 0.0f;
+#pragma unroll
+        for (int k = 0; k < QK_NVFP4_SUB; ++k) {
+            const float v = vals_raw[k];
+            const uint8_t q = ggml_cuda_float_to_fp4_e2m1(v, test_inv_scale);
+            const float err_diff = fabsf(v) - fabsf(kvalues_mxfp4[q & 0x7]) * test_scale;
+            cur_err = fmaf(err_diff, err_diff, cur_err);
+        }
+
+        if (cur_err < best_err) {
+            best_err = cur_err;
+            fp8_code = test_code;
+            subblock_scale = test_scale;
+        }
+    }
+
+    const float inv_scale = subblock_scale > 0.0f ? 0.5f / subblock_scale : 0.0f;
+    uint32_t q0 = 0;
+    uint32_t q1 = 0;
+#pragma unroll // this is faster than the previous __nv_fp4x4_e2m1
+    for (int k = 0; k < QK_NVFP4_SUB / 4; ++k) {
+        q0 |= (uint32_t) ggml_cuda_float_to_fp4_e2m1(vals_raw[k +  0], inv_scale) << (8 * k);
+        q0 |= (uint32_t) ggml_cuda_float_to_fp4_e2m1(vals_raw[k +  8], inv_scale) << (8 * k + 4);
+        q1 |= (uint32_t) ggml_cuda_float_to_fp4_e2m1(vals_raw[k +  4], inv_scale) << (8 * k);
+        q1 |= (uint32_t) ggml_cuda_float_to_fp4_e2m1(vals_raw[k + 12], inv_scale) << (8 * k + 4);
+    }
+
+    uint32_t * yqs = reinterpret_cast<uint32_t *>(yb->qs);
+    yqs[2 * sub + 0] = q0;
+    yqs[2 * sub + 1] = q1;
+    reinterpret_cast<uint8_t *>(yb->d4)[sub] = fp8_code;
+#else
+    NO_DEVICE_CODE; // This is for Blackwell NVFP4 activations only.
+#endif // defined(BLACKWELL_MMA_AVAILABLE)
+
+}
+
 // quantize values in the format mxfp4 is stored which is interleaved nibbles
 // i.e. a block a0-a31 is represented as a0a16,a1a17 ...a15a31
 static __global__ void quantize_mmq_mxfp4(const float * __restrict__ x,
@@ -316,28 +412,32 @@ void quantize_mmq_q8_1_cuda(
     }
 }
 
-void quantize_mmq_mxfp4_cuda(const float *                    x,
-                             const int32_t *                  ids,
-                             void *                           vy,
-                             [[maybe_unused]] const ggml_type type_src0,
-                             const int64_t                    ne00,
-                             const int64_t                    s01,
-                             const int64_t                    s02,
-                             const int64_t                    s03,
-                             const int64_t                    ne0,
-                             const int64_t                    ne1,
-                             const int64_t                    ne2,
-                             const int64_t                    ne3,
-                             cudaStream_t                     stream) {
-    GGML_ASSERT(ne0 % (2 * QK_MXFP4) == 0);
+void quantize_mmq_fp4_cuda(
+        const float * x, const int32_t * ids, void * vy, const ggml_type type_src0,
+        const int64_t ne00, const int64_t s01, const int64_t s02, const int64_t s03,
+        const int64_t ne0, const int64_t ne1, const int64_t ne2, const int64_t ne3, cudaStream_t stream) {
+    GGML_ASSERT(type_src0 == GGML_TYPE_MXFP4 || type_src0 == GGML_TYPE_NVFP4);
+    GGML_ASSERT(ne0 > 0);
 
-    constexpr int nwarps = 8;
-    constexpr int vals_per_warp  = 2 * QK_MXFP4;
-    constexpr int vals_per_block = nwarps * vals_per_warp;
+    if (type_src0 == GGML_TYPE_NVFP4) {
+        GGML_ASSERT(ne00 % QK_NVFP4 == 0);
+        constexpr int nvfp4_block_size = 128;
+        const int64_t block_num_y = (ne0 + QK_NVFP4_SUB * nvfp4_block_size - 1) / (QK_NVFP4_SUB * nvfp4_block_size);
+        const dim3 block_size(nvfp4_block_size, 1, 1);
+        const dim3 num_blocks(ne1, block_num_y, ne2 * ne3);
+        quantize_mmq_nvfp4<<<num_blocks, block_size, 0, stream>>>(
+            x, ids, vy, ne00, s01, s02, s03, ne0, ne1, ne2);
+    } else {
+        GGML_ASSERT(ne0 % (2 * QK_MXFP4) == 0);
 
-    const int64_t block_num_y = (ne0 + vals_per_block - 1) / vals_per_block;
-    const dim3    num_blocks(ne1, block_num_y, ne2 * ne3);
-    const dim3    block_size(WARP_SIZE, nwarps, 1);
+        constexpr int nwarps = 8;
+        constexpr int vals_per_warp  = 2 * QK_MXFP4;
+        constexpr int vals_per_block = nwarps * vals_per_warp;
 
-    quantize_mmq_mxfp4<<<num_blocks, block_size, 0, stream>>>(x, ids, vy, ne00, s01, s02, s03, ne0, ne1, ne2);
+        const int64_t block_num_y = (ne0 + vals_per_block - 1) / vals_per_block;
+        const dim3    num_blocks(ne1, block_num_y, ne2 * ne3);
+        const dim3    block_size(WARP_SIZE, nwarps, 1);
+
+        quantize_mmq_mxfp4<<<num_blocks, block_size, 0, stream>>>(x, ids, vy, ne00, s01, s02, s03, ne0, ne1, ne2);
+    }
 }

ggml/src/ggml-cuda/quantize.cuh

@@ -26,7 +26,7 @@ void quantize_mmq_q8_1_cuda(
         ggml_type type_src0, int64_t ne00, int64_t s01, int64_t s02, int64_t s03,
         int64_t ne0, int64_t ne1, int64_t ne2, int64_t ne3, cudaStream_t stream);
 
-void quantize_mmq_mxfp4_cuda(const float *   x,
+void quantize_mmq_fp4_cuda(const float *   x,
                              const int32_t * ids,
                              void *          vy,
                              ggml_type       type_src0,

ggml/src/ggml-cuda/ssm-conv.cu

@@ -3,6 +3,7 @@
 
 template <bool apply_silu, size_t split_d_inner, size_t d_conv>
 static __global__ void ssm_conv_f32(const float * __restrict__ src0, const float * __restrict__ src1,
+                                    const float * __restrict__ bias,
                                     const int src0_nb0, const int src0_nb1, const int src0_nb2, const int src1_nb1,
                                     float * __restrict__ dst, const int dst_nb0, const int dst_nb1, const int dst_nb2,
                                     const int64_t n_t) {
@@ -27,6 +28,8 @@ static __global__ void ssm_conv_f32(const float * __restrict__ src0, const float
         w[j] = w_block[tid * stride_w + j];
     }
 
+    float b = bias != nullptr ? bias[bidy * split_d_inner + tid] : 0.0f;
+
     for (int64_t i = 0; i < n_t; i++) {
         float sumf = 0.0f;
 
@@ -42,12 +45,14 @@ static __global__ void ssm_conv_f32(const float * __restrict__ src0, const float
         for (size_t j = 0; j < d_conv; j++) {
             sumf += x[(i + j) % d_conv] * w[j];
         }
+        sumf += b;
         y_block[i * stride_y + tid] = apply_silu ? ggml_cuda_op_silu_single(sumf) : sumf;
     }
 }
 
 template <bool apply_silu, size_t split_d_inner, size_t d_conv, int64_t split_n_t>
 static __global__ void ssm_conv_long_token_f32(const float * __restrict__ src0, const float * __restrict__ src1,
+                                               const float * __restrict__ bias,
                                                const int src0_nb0, const int src0_nb1, const int src0_nb2,
                                                const int src1_nb1, float * __restrict__ dst, const int dst_nb0,
                                                const int dst_nb1, const int dst_nb2, const int64_t n_t) {
@@ -97,6 +102,8 @@ static __global__ void ssm_conv_long_token_f32(const float * __restrict__ src0,
         w[j] = w_block[tid * stride_w + j];
     }
 
+    float b = bias != nullptr ? bias[bidy * split_d_inner + tid] : 0.0f;
+
     // Compute from shared memory
     for (int64_t i = 0; i < local_n_t; i++) {
         float sumf = 0.0f;
@@ -104,12 +111,13 @@ static __global__ void ssm_conv_long_token_f32(const float * __restrict__ src0,
         for (size_t j = 0; j < d_conv; j++) {
             sumf += smem[tid * n_cols + i + j] * w[j];
         }
+        sumf += b;
         y_block[i * stride_y + tid] = apply_silu ? ggml_cuda_op_silu_single(sumf) : sumf;
     }
 }
 
 template <bool apply_silu>
-static void ssm_conv_f32_cuda(const float * src0, const float * src1, const int src0_nb0, const int src0_nb1,
+static void ssm_conv_f32_cuda(const float * src0, const float * src1, const float * bias, const int src0_nb0, const int src0_nb1,
                               const int src0_nb2, const int src1_nb1, float * dst, const int dst_nb0, const int dst_nb1,
                               const int dst_nb2, const int64_t nc, const int64_t nr, const int64_t n_t,
                               const int64_t n_s, cudaStream_t stream) {
@@ -120,14 +128,14 @@ static void ssm_conv_f32_cuda(const float * src0, const float * src1, const int
         constexpr int kNC = decltype(NC)::value;
         if (n_t <= 32) {
             const dim3 blocks(n_s, (nr + threads - 1) / threads, 1);
-            ssm_conv_f32<apply_silu, threads, kNC><<<blocks, threads, 0, stream>>>(src0, src1, src0_nb0, src0_nb1, src0_nb2, src1_nb1,
+            ssm_conv_f32<apply_silu, threads, kNC><<<blocks, threads, 0, stream>>>(src0, src1, bias, src0_nb0, src0_nb1, src0_nb2, src1_nb1,
                                                                        dst, dst_nb0, dst_nb1, dst_nb2, n_t);
         } else {
             const int64_t split_n_t = 32;
             dim3          blocks(n_s, (nr + threads - 1) / threads, (n_t + split_n_t - 1) / split_n_t);
             const size_t  smem_size = threads * (kNC - 1 + split_n_t) * sizeof(float);
             ssm_conv_long_token_f32<apply_silu, threads, kNC, split_n_t><<<blocks, threads, smem_size, stream>>>(
-                src0, src1, src0_nb0, src0_nb1, src0_nb2, src1_nb1, dst, dst_nb0, dst_nb1, dst_nb2, n_t);
+                src0, src1, bias, src0_nb0, src0_nb1, src0_nb2, src1_nb1, dst, dst_nb0, dst_nb1, dst_nb2, n_t);
         }
     };
 
@@ -140,11 +148,18 @@ static void ssm_conv_f32_cuda(const float * src0, const float * src1, const int
     }
 }
 
-void ggml_cuda_op_ssm_conv(ggml_backend_cuda_context & ctx, ggml_tensor * dst, ggml_tensor * silu_dst) {
+void ggml_cuda_op_ssm_conv(ggml_backend_cuda_context & ctx, ggml_tensor * dst, ggml_tensor * bias_add_node, ggml_tensor * silu_dst) {
     const struct ggml_tensor * src0 = dst->src[0];  // conv_x
     const struct ggml_tensor * src1 = dst->src[1];  // conv1d.weight
+    const bool fuse_bias = bias_add_node != nullptr;
     const bool fuse_silu = silu_dst != nullptr;
 
+    // bias always comes with silu.
+    GGML_ASSERT(!fuse_bias || fuse_silu);
+
+    // The bias (when fused) is the non-conv operand of the ADD node.
+    const struct ggml_tensor * bias = fuse_bias ? (bias_add_node->src[0] == dst ? bias_add_node->src[1] : bias_add_node->src[0]) : nullptr;
+
     // When fusing, write to silu_dst (the node downstream references).
     const struct ggml_tensor * out = fuse_silu ? silu_dst : dst;
 
@@ -160,16 +175,23 @@ void ggml_cuda_op_ssm_conv(ggml_backend_cuda_context & ctx, ggml_tensor * dst, g
 
     const float * src0_d = (const float *) src0->data;
     const float * src1_d = (const float *) src1->data;
+    const float * bias_d = fuse_bias ? (const float *) bias->data : nullptr;
     float *       dst_d  = (float *) out->data;
     cudaStream_t  stream = ctx.stream();
 
     GGML_ASSERT(src0->type == GGML_TYPE_F32);
     GGML_ASSERT(out->type == GGML_TYPE_F32);
+    if (fuse_bias) {
+        GGML_ASSERT(bias->type == GGML_TYPE_F32);
+        GGML_ASSERT(ggml_is_contiguous(bias));
+        GGML_ASSERT(ggml_nelements(bias) == nr);
+    }
+
     if (fuse_silu) {
-        ssm_conv_f32_cuda<true>(src0_d, src1_d, src0->nb[0], src0->nb[1], src0->nb[2], src1->nb[1], dst_d, out->nb[0], out->nb[1],
+        ssm_conv_f32_cuda<true>(src0_d, src1_d, bias_d, src0->nb[0], src0->nb[1], src0->nb[2], src1->nb[1], dst_d, out->nb[0], out->nb[1],
                           out->nb[2], nc, nr, n_t, n_s, stream);
     } else {
-        ssm_conv_f32_cuda<false>(src0_d, src1_d, src0->nb[0], src0->nb[1], src0->nb[2], src1->nb[1], dst_d, out->nb[0], out->nb[1],
+        ssm_conv_f32_cuda<false>(src0_d, src1_d, bias_d, src0->nb[0], src0->nb[1], src0->nb[2], src1->nb[1], dst_d, out->nb[0], out->nb[1],
                           out->nb[2], nc, nr, n_t, n_s, stream);
     }
 }

ggml/src/ggml-cuda/ssm-conv.cuh

@@ -1,3 +1,3 @@
 #include "common.cuh"
 
-void ggml_cuda_op_ssm_conv(ggml_backend_cuda_context & ctx, ggml_tensor * dst, ggml_tensor * silu_dst = nullptr);
+void ggml_cuda_op_ssm_conv(ggml_backend_cuda_context & ctx, ggml_tensor * dst, ggml_tensor * bias_add_node = nullptr, ggml_tensor * silu_dst = nullptr);

ggml/src/ggml-cuda/template-instances/fattn-mma-f16-instance-ncols1_1-ncols2_32.cu

@@ -2,4 +2,5 @@
 
 #include "../fattn-mma-f16.cuh"
 
+DECL_FATTN_MMA_F16_CASE(320, 256, 1, 32);
 DECL_FATTN_MMA_F16_CASE(576, 512, 1, 32);

ggml/src/ggml-cuda/template-instances/fattn-mma-f16-instance-ncols1_2-ncols2_32.cu

@@ -2,4 +2,5 @@
 
 #include "../fattn-mma-f16.cuh"
 
+DECL_FATTN_MMA_F16_CASE(320, 256, 2, 32);
 DECL_FATTN_MMA_F16_CASE(576, 512, 2, 32);

ggml/src/ggml-cuda/template-instances/fattn-tile-instance-dkq320-dv256.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate_cu_files.py, do not edit manually.
+
+#include "../fattn-tile.cuh"
+
+DECL_FATTN_TILE_CASE(320, 256);

ggml/src/ggml-cuda/template-instances/generate_cu_files.py

@@ -3,7 +3,7 @@
 from glob import glob
 import os
 
-HEAD_SIZES_KQ = [40, 64, 72, 80, 96, 112, 128, 256, 512, 576]
+HEAD_SIZES_KQ = [40, 64, 72, 80, 96, 112, 128, 256, 320, 512, 576]
 
 TYPES_KV = ["GGML_TYPE_F16", "GGML_TYPE_Q4_0", "GGML_TYPE_Q4_1", "GGML_TYPE_Q5_0", "GGML_TYPE_Q5_1", "GGML_TYPE_Q8_0", "GGML_TYPE_BF16"]
 
@@ -62,7 +62,7 @@ for filename in glob("*.cu"):
     os.remove(filename)
 
 for head_size_kq in HEAD_SIZES_KQ:
-    head_size_v = head_size_kq if head_size_kq != 576 else 512
+    head_size_v = 256 if head_size_kq == 320 else (head_size_kq if head_size_kq != 576 else 512)
     with open(f"fattn-tile-instance-dkq{head_size_kq}-dv{head_size_v}.cu", "w") as f:
         f.write(SOURCE_FATTN_TILE.format(head_size_kq=head_size_kq, head_size_v=head_size_v))
 
@@ -84,13 +84,16 @@ for ncols in [8, 16, 32, 64]:
                     continue
                 if head_size_kq == 72:
                     continue
-                if head_size_kq == 512 and ncols2 not in (4, 8):
+                # Skip compilation of unused ncols2 values for niche head sizes:
+                if head_size_kq == 320 and ncols2 != 32: # Mistral Small 4
                     continue
-                if head_size_kq != 576 and ncols2 in (16, 32):
+                if head_size_kq == 512 and ncols2 not in (4, 8): # Gemma 4
                     continue
-                if head_size_kq == 576 and ncols2 not in (4, 16, 32):
+                if head_size_kq == 576 and ncols2 not in (4, 16, 32): # Deepseek, GLM 4.7 Flash
                     continue
-                head_size_v = head_size_kq if head_size_kq != 576 else 512
+                if head_size_kq not in (320, 576) and ncols2 in (16, 32):
+                    continue
+                head_size_v = 256 if head_size_kq == 320 else (head_size_kq if head_size_kq != 576 else 512)
                 f.write(SOURCE_FATTN_MMA_CASE.format(ncols1=ncols1, ncols2=ncols2, head_size_kq=head_size_kq, head_size_v=head_size_v))
 
 for type in TYPES_MMQ:

ggml/src/ggml-cuda/vendors/hip.h

@@ -55,6 +55,7 @@
 #define cudaDeviceDisablePeerAccess hipDeviceDisablePeerAccess
 #define cudaDeviceEnablePeerAccess hipDeviceEnablePeerAccess
 #define cudaDeviceGetAttribute hipDeviceGetAttribute
+#define cudaDeviceGetPCIBusId hipDeviceGetPCIBusId
 #define cudaDeviceProp hipDeviceProp_t
 #define cudaDeviceSynchronize hipDeviceSynchronize
 #define cudaError_t hipError_t

ggml/src/ggml-cuda/vendors/musa.h

@@ -39,6 +39,7 @@
 #define cudaDeviceCanAccessPeer musaDeviceCanAccessPeer
 #define cudaDeviceDisablePeerAccess musaDeviceDisablePeerAccess
 #define cudaDeviceEnablePeerAccess musaDeviceEnablePeerAccess
+#define cudaDeviceGetPCIBusId musaDeviceGetPCIBusId
 #define cudaDeviceProp musaDeviceProp
 #define cudaDeviceSynchronize musaDeviceSynchronize
 #define cudaError_t musaError_t

ggml/src/ggml-hexagon/CMakeLists.txt

@@ -22,7 +22,8 @@ message(STATUS "hexagon: using ${HEXAGON_SDK_ROOT} and ${HEXAGON_TOOLS_ROOT} for
 include(${HEXAGON_SDK_ROOT}/build/cmake/hexagon_fun.cmake)
 include(ExternalProject)
 
-option(GGML_HEXAGON_HTP_DEBUG "ggml-hexagon: enable HTP debug output" OFF)
+option(GGML_HEXAGON_HTP_DEBUG  "ggml-hexagon: enable HTP debug output" OFF)
+option(GGML_HEXAGON_FA_EXP2_HF "ggml-hexagon: use FP16 exp2 polynomial in FA softmax instead of F32 exp round-trip" OFF)
 set(GGML_HEXAGON_HTP_CERT  "$ENV{HEXAGON_HTP_CERT}" CACHE PATH "ggml-hexagon: enable HTP library signing using certificate")
 set(GGML_HEXAGON_FP32_QUANTIZE_GROUP_SIZE 128 CACHE STRING "ggml-hexagon: quantize group size (32, 64, or 128)")
 

ggml/src/ggml-hexagon/ggml-hexagon.cpp

@@ -48,14 +48,16 @@ using intvec  = std::vector<int>;
 using uintvec = std::vector<unsigned int>;
 using u32vec  = std::vector<uint32_t>;
 
-static size_t opt_ndev         = 1;
-static size_t opt_nhvx         = 0; // use all
-static int    opt_arch         = 0; // autodetect
-static int    opt_etm          = 0;
-static int    opt_verbose      = 0;
-static int    opt_profile      = 0; // profiling mode (0-disabled, 1-basic, 2-pmu)
-static int    opt_hostbuf      = 1; // hostbuf ON by default
-static int    opt_use_hmx      = 1; // when set, enable HMX; when 0, use HVX only
+static int    opt_arch    = 0; // autodetect
+static size_t opt_ndev    = 1;
+static size_t opt_nhvx    = 0; // use all
+static int    opt_use_hmx = 1; // when set, enable HMX; when 0, use HVX only
+static size_t opt_vmem    = HTP_OP_MAX_VMEM_DEFAULT;  // max available va space for buffer mappings
+static size_t opt_mbuf    = 1ul * 1024 * 1024 * 1024; // max buffer size
+static int    opt_etm     = 0;
+static int    opt_verbose = 0;
+static int    opt_profile = 0; // profiling mode (0-disabled, 1-basic, 2-pmu)
+static int    opt_hostbuf = 1; // hostbuf ON by default
 
 // Default PMU events, if profiling with PMU (mode=2) is enabled
 // See https://docs.qualcomm.com/doc/80-N2040-60/topic/pmu-events.html
@@ -66,6 +68,7 @@ static u32vec opt_pmu_evt { 0x3, 0x111, 0x100, 0x105, 0x240, 0x256, 0x7D, 0x8C }
 static int opt_opstage  = HTP_OPSTAGE_QUEUE | HTP_OPSTAGE_COMPUTE;
 static int opt_opbatch  = 1024; // max number of ops in a batch
 static int opt_opqueue  = 16;   // max number of pending batches
+
 static std::regex* opt_opfilter = NULL; // regex of ops to not claim
 
 #define HEX_VERBOSE(...) \
@@ -110,7 +113,7 @@ static void ggml_hexagon_dump_op_supp(const std::string &sess_name, const struct
     if (!opt_verbose) return;
 
     op_desc desc(op);
-    GGML_LOG_DEBUG("ggml-hex: %s supports-op %s : %s : %s : %s : %s : %s : %s\n", sess_name.c_str(),
+    GGML_LOG_DEBUG("ggml-hex: %s supports-op %s: %s : %s : %s : %s : %s : %s\n", sess_name.c_str(),
                 ggml_op_desc(op), desc.names, desc.dims, desc.types, desc.strides, desc.buffs, supp ? "yes" : "no");
 }
 
@@ -118,8 +121,6 @@ static void ggml_hexagon_dump_op_prof(const std::string &sess_name, const ggml_t
                                       uint32_t op_usec, uint32_t op_cycles, const uint32_t pmu[]) {
     if (!opt_profile) return;
 
-    op_desc desc(op);
-
     char pmu_str[256] = "";
     if (opt_profile > 1) {
         static_assert(HTP_PROF_PMU_NCNT == 8, "current implementation assumes 8 PMU counters");
@@ -127,6 +128,7 @@ static void ggml_hexagon_dump_op_prof(const std::string &sess_name, const ggml_t
                 pmu[0], pmu[1], pmu[2], pmu[3], pmu[4], pmu[5], pmu[6], pmu[7]);
     }
 
+    op_desc desc(op);
     GGML_LOG_DEBUG("ggml-hex: %s profile-op %s: %s : %s : %s : %s : usec %u cycles %u%s\n", sess_name.c_str(),
             ggml_op_desc(op), desc.names, desc.dims, desc.types, desc.strides, op_usec, op_cycles, pmu_str);
 }
@@ -191,33 +193,30 @@ struct ggml_hexagon_shared_buffer {
     bool                   mapped;
     bool                   pinned;
 
-    void mmap(bool pinned = false) {
-        int err = fastrpc_mmap(sess->domain_id, this->fd, (void *) this->base, 0, this->size, FASTRPC_MAP_FD_DELAYED);
+    void mmap() {
+        fastrpc_map_flags flags = this->pinned ? FASTRPC_MAP_FD : FASTRPC_MAP_FD_DELAYED;
+
+        int err = fastrpc_mmap(sess->domain_id, this->fd, (void *) this->base, 0, this->size, flags);
         if (err != 0) {
             GGML_LOG_ERROR("ggml-hex: %s buffer mapping failed : domain_id %d size %zu fd %d error 0x%08x\n", sess->c_name(),
                     sess->domain_id, this->size, this->fd, (unsigned) err);
             throw std::runtime_error("ggml-hex: fastrpc_mmap failed (see log for details)");
         }
 
-        if (pinned) {
-            err = htp_iface_mmap(sess->handle, this->fd, this->size, pinned);
-            if (err != 0) {
-                GGML_LOG_ERROR("ggml-hex: %s buffer pinning failed : domain_id %d size %zu fd %d error 0x%08x\n", sess->c_name(),
-                        sess->domain_id, this->size, this->fd, (unsigned) err);
-                throw std::runtime_error("ggml-hex: htp_iface_mmap failed (see log for details)");
-            }
-        }
-
-        this->mapped = true;
-        this->pinned = pinned;
         HEX_VERBOSE("ggml-hex: %s mapped buffer: base %p size %zu fd %d pinned %u\n",
                 sess->c_name(), (void *) this->base, this->size, this->fd, pinned);
+
+        this->mapped = true;
     }
 
     void unmap() {
         if (!this->mapped) return;
 
-        htp_iface_munmap(sess->handle, this->fd);
+        if (!this->pinned) {
+            // HTP might still hold a reference, tell it drop it
+            htp_iface_munmap(sess->handle, this->fd);
+        }
+
         fastrpc_munmap(sess->domain_id, this->fd, (void *) this->base, this->size);
 
         HEX_VERBOSE("ggml-hex: %s unmapped buffer: base %p size %zu fd %d\n", sess->c_name(),
@@ -227,7 +226,7 @@ struct ggml_hexagon_shared_buffer {
         this->fd     = -1;
     }
 
-    void alloc(size_t size, bool pinned = false) {
+    void alloc(size_t size) {
         if (this->base) return;
 
         this->base = (uint8_t *) rpcmem_alloc2(RPCMEM_HEAP_ID_SYSTEM, RPCMEM_DEFAULT_FLAGS, size);
@@ -245,8 +244,7 @@ struct ggml_hexagon_shared_buffer {
 
         HEX_VERBOSE("ggml-hex: %s allocated buffer: base %p size %zu fd %d pinned %d\n", sess->c_name(),
                     (void *) this->base, this->size, this->fd, (int) pinned);
-
-        mmap(pinned);
+        mmap();
     }
 
     void free() {
@@ -262,15 +260,14 @@ struct ggml_hexagon_shared_buffer {
     }
 
     ggml_hexagon_shared_buffer(ggml_hexagon_session * sess, size_t size, bool pinned = false) {
-        size += 4 * 1024;  // extra page for padding
-
         this->sess   = sess;
         this->size   = 0;
         this->base   = nullptr;
         this->fd     = -1;
         this->mapped = false;
+        this->pinned = pinned;
 
-        alloc(size, pinned);
+        alloc(size);
     }
 
     ~ggml_hexagon_shared_buffer() {
@@ -1475,6 +1472,7 @@ static ggml_backend_buffer_t ggml_backend_hexagon_buffer_type_alloc_buffer(
             ggml_backend_buffer_type_t buffer_type, size_t size) {
     auto sess = static_cast<ggml_backend_hexagon_buffer_type_context *>(buffer_type->context)->sess;
     try {
+        size += 4 * 1024;  // guard page
         ggml_hexagon_shared_buffer * sbuf = new ggml_hexagon_shared_buffer(sess, size);
         return ggml_backend_buffer_init(buffer_type, ggml_backend_hexagon_buffer_interface, sbuf, size);
     } catch (const std::exception & exc) {
@@ -1487,6 +1485,7 @@ static ggml_backend_buffer_t ggml_backend_hexagon_repack_buffer_type_alloc_buffe
             ggml_backend_buffer_type_t buffer_type, size_t size) {
     auto sess = static_cast<ggml_backend_hexagon_buffer_type_context *>(buffer_type->context)->sess;
     try {
+        size += 4 * 1024;  // guard page
         ggml_hexagon_shared_buffer * sbuf = new ggml_hexagon_shared_buffer(sess, size);
         return ggml_backend_buffer_init(buffer_type, ggml_backend_hexagon_buffer_interface, sbuf, size);
     } catch (const std::exception & exc) {
@@ -1505,7 +1504,7 @@ static size_t ggml_backend_hexagon_buffer_type_get_alloc_size(ggml_backend_buffe
 }
 
 static size_t ggml_backend_hexagon_buffer_type_get_max_size(ggml_backend_buffer_type_t buffer_type) {
-    return 1UL * 1024 * 1024 * 1024;  // 1GB per buffer
+    return opt_mbuf; // typically 1GB per buffer
     GGML_UNUSED(buffer_type);
 }
 
@@ -1573,14 +1572,14 @@ struct ggml_hexagon_opbatch {
         d_map.clear();
     }
 
-    ggml_hexagon_opbatch(ggml_hexagon_session *sess, size_t batch_size) {
+    ggml_hexagon_opbatch(ggml_hexagon_session *sess, size_t batch_size, size_t max_vmem) {
         this->sess = sess;
 
         n_bufs_max = HTP_OP_MAX_BUFS;
         n_ops_max  = batch_size;
         n_tens_max = n_ops_max + n_ops_max * HTP_OP_MAX_INPUTS;
 
-        b_vmem_max = HTP_OP_MAX_VMEM;
+        b_vmem_max = max_vmem;
 
         ops.resize(n_ops_max);
 
@@ -1592,6 +1591,9 @@ struct ggml_hexagon_opbatch {
         t_map.reserve(n_tens_max);
         d_map.reserve(n_tens_max);
 
+        GGML_LOG_INFO("ggml-hex: %s op batching: n-bufs %u n-tensors %u n-ops %u vmem %zu\n",
+                sess->c_name(), n_bufs_max, n_tens_max, n_ops_max, b_vmem_max);
+
         reset();
     }
 
@@ -1925,6 +1927,8 @@ void ggml_hexagon_session::flush_batch() {
     // Bump pending flag (cleared in the session::flush once we get the response)
     this->op_pending++;  // atomic inc
 
+    HEX_VERBOSE("ggml-hex: %s queue-opbatch: %p size %u\n", this->c_name(), dbuf.ptr, dbuf.size);
+
     int err = dspqueue_write(this->queue, 0, 1, &dbuf, sizeof(req), (const uint8_t*) &req, DSPQUEUE_TIMEOUT);
     if (err != 0) {
         GGML_ABORT("ggml-hex: %s dspqueue_write failed: 0x%08x\n", this->c_name(), (unsigned) err);
@@ -1944,6 +1948,35 @@ void ggml_hexagon_session::flush(bool all) {
     flush_pending(all);
 }
 
+static size_t ggml_hexagon_measure_max_vmem(ggml_hexagon_session *sess) {
+    // Allocate a bunch pinned buffers till failure.
+    // This is kind of expensive but handy for figuring out exactly how much we can mmap on a specific device.
+    // Typically we're going to allocate all/most of these buffers anyway for the model weights.
+
+    std::vector<ggml_hexagon_shared_buffer *> sbufs;
+
+    const size_t MiB = 1024 * 1024;
+    const size_t GiB = MiB  * 1024;
+
+    size_t vmem = 0;
+    size_t step = 256u * MiB;
+
+    try {
+        sbufs.push_back(new ggml_hexagon_shared_buffer(sess, GiB, true)); vmem += GiB;
+        sbufs.push_back(new ggml_hexagon_shared_buffer(sess, GiB, true)); vmem += GiB;
+        sbufs.push_back(new ggml_hexagon_shared_buffer(sess, GiB, true)); vmem += GiB;
+
+        while (1) {
+            sbufs.push_back(new ggml_hexagon_shared_buffer(sess, step, true));
+            vmem += step;
+        }
+    } catch (...) { }
+
+    for (auto b : sbufs) { delete b; }
+
+    return vmem - step; // backoff to account for overhead from internal mappings
+}
+
 void ggml_hexagon_session::allocate(int dev_id) noexcept(false) {
     this->valid_session = false;
     this->valid_handle  = false;
@@ -1957,7 +1990,7 @@ void ggml_hexagon_session::allocate(int dev_id) noexcept(false) {
 
     this->op_pending  = 0;
 
-    GGML_LOG_INFO("ggml-hex: allocating new session: %s\n", this->name.c_str());
+    GGML_LOG_DEBUG("ggml-hex: %s allocating new session\n", this->name.c_str());
 
     domain * my_domain = get_domain(this->domain_id);
     if (my_domain == NULL) {
@@ -2033,9 +2066,6 @@ void ggml_hexagon_session::allocate(int dev_id) noexcept(false) {
 
     this->valid_handle = true;
 
-    GGML_LOG_INFO("ggml-hex: new session: %s : session-id %d domain-id %d uri %s handle 0x%lx\n", this->name.c_str(),
-                  this->session_id, this->domain_id, session_uri, (unsigned long) this->handle);
-
     // Enable FastRPC QoS mode
     {
         struct remote_rpc_control_latency l;
@@ -2047,6 +2077,9 @@ void ggml_hexagon_session::allocate(int dev_id) noexcept(false) {
         }
     }
 
+    GGML_LOG_INFO("ggml-hex: %s new session : session-id %d domain-id %d uri %s handle 0x%lx\n", this->c_name(),
+                  this->session_id, this->domain_id, session_uri, (unsigned long) this->handle);
+
     const size_t req_q_size = (sizeof(htp_opbatch_req) * opt_opqueue * 2) + 1024;
     const size_t rsp_q_size = (sizeof(htp_opbatch_rsp) * opt_opqueue * 2) + 1024;
 
@@ -2091,13 +2124,19 @@ void ggml_hexagon_session::allocate(int dev_id) noexcept(false) {
     }
 
     // Allocate buffers and state for op batching
-    this->op_batch = new ggml_hexagon_opbatch(this, opt_opbatch);
     this->op_queue = new ggml_hexagon_opqueue(this, opt_opbatch, opt_opqueue);
 
-    // Start processing op batch requests
-    err = htp_iface_start(this->handle, dev_id, this->queue_id, opt_nhvx, opt_use_hmx);
+    if (!opt_vmem) {
+        opt_vmem = ggml_hexagon_measure_max_vmem(this);
+        GGML_LOG_INFO("ggml-hex: %s measured max vmem %zu\n", this->c_name(), opt_vmem);
+    }
+
+    this->op_batch = new ggml_hexagon_opbatch(this, opt_opbatch, opt_vmem);
+
+    // Start dspqueue/opbatch processing
+    err = htp_iface_start(this->handle, dev_id, this->queue_id, opt_nhvx, opt_use_hmx, opt_vmem);
     if (err != 0) {
-        GGML_LOG_ERROR("ggml-hex: failed to start session: 0x%08x\n", (unsigned) err);
+        GGML_LOG_ERROR("ggml-hex: %s failed to start session: 0x%08x\n", this->c_name(), (unsigned) err);
         throw std::runtime_error("ggml-hex: iface start failed (see log for details)");
     }
     this->valid_iface = true;
@@ -2108,17 +2147,17 @@ void ggml_hexagon_session::release() noexcept(true) {
 
     int err;
 
-    delete this->op_batch;
-    delete this->op_queue;
-
-    // Stop the DSP-side service and close the queue
     if (this->valid_iface) {
+        // Stop dspqueue/opbatch processing
         err = htp_iface_stop(this->handle);
         if (err != 0) {
             GGML_ABORT("ggml-hex: htp_iface_stop failed: 0x%08x\n", (unsigned) err);
         }
     }
 
+    delete this->op_batch;
+    delete this->op_queue;
+
     if (opt_etm) {
         err = htp_iface_etm(this->handle, 0);
         if (err != 0) {
@@ -2215,8 +2254,7 @@ static bool ggml_hexagon_supported_flash_attn_ext(const struct ggml_hexagon_sess
         return false;
     }
 
-    if (dst->ne[2] != 1 || dst->ne[3] != 1) {
-        // FA during prompt still needs work
+    if (dst->ne[3] != 1) {
         return false;
     }
 
@@ -2382,8 +2420,8 @@ static bool ggml_hexagon_supported_unary(const struct ggml_hexagon_session * ses
         return false;
     }
 
-    // TODO: add support for non-contigiuos tensors
-    if (!ggml_is_contiguous(src0) || !ggml_is_contiguous(dst)) {
+    // TODO: add support for non-contiguous elements within a row
+    if (!ggml_is_contiguous_rows(src0) || !ggml_is_contiguous_rows(dst)) {
         return false;
     }
 
@@ -2997,8 +3035,8 @@ static struct ggml_backend_i hexagon_backend_i = {
     /* .free                    = */ ggml_backend_hexagon_free,
     /* .set_tensor_async        = */ NULL,
     /* .get_tensor_async        = */ NULL,
-    /* .get_tensor_2d_async     = */ NULL,
     /* .set_tensor_2d_async     = */ NULL,
+    /* .get_tensor_2d_async     = */ NULL,
     /* .cpy_tensor_async        = */ NULL,
     /* .synchronize             = */ ggml_backend_hexagon_synchronize,
     /* .graph_plan_create       = */ NULL,
@@ -3380,21 +3418,6 @@ struct ggml_hexagon_registry {
 ggml_hexagon_registry::ggml_hexagon_registry(ggml_backend_reg_t reg) {
     GGML_LOG_INFO("ggml-hex: Hexagon backend (experimental) : allocating new registry : ndev %zu\n", opt_ndev);
 
-    if (!opt_arch) {
-        int err = get_hex_arch_ver(CDSP_DOMAIN_ID, &opt_arch);
-        if (err != 0) {
-            GGML_LOG_ERROR("ggml-hex: failed to query HTP version (err %d) defaulting to v73\n", err);
-            opt_arch = 73;
-        }
-    }
-
-#if defined(__ANDROID__)
-    if (opt_arch < 75) {
-        opt_ndev = 1;
-        GGML_LOG_WARN("ggml-hex: forcing ndev to 1 for SoCs archs lower than v75.\n");
-    }
-#endif
-
     GGML_LOG_INFO("ggml-hex: Hexagon Arch version v%d\n", opt_arch);
 
     // Create devices / sessions
@@ -3480,32 +3503,67 @@ static void ggml_hexagon_init(ggml_backend_reg * reg) {
     static_assert((unsigned int) HTP_TYPE_IQ4_NL == (unsigned int) GGML_TYPE_IQ4_NL,
                   "please update hexagon_type to match ggml_type");
 
-    const char * str_verbose = getenv("GGML_HEXAGON_VERBOSE");
-    const char * str_hostbuf = getenv("GGML_HEXAGON_HOSTBUF");
-    const char * str_opstage = getenv("GGML_HEXAGON_OPSTAGE");
-    const char * str_opbatch = getenv("GGML_HEXAGON_OPBATCH");
-    const char * str_opqueue = getenv("GGML_HEXAGON_OPQUEUE");
-    const char * str_opfilter= getenv("GGML_HEXAGON_OPFILTER");
-    const char * str_profile = getenv("GGML_HEXAGON_PROFILE");
-    const char * str_etm     = getenv("GGML_HEXAGON_ETM");
-    const char * str_nhvx    = getenv("GGML_HEXAGON_NHVX");
-    const char * str_use_hmx = getenv("GGML_HEXAGON_USE_HMX");
-    const char * str_ndev    = getenv("GGML_HEXAGON_NDEV");
-    const char * str_arch    = getenv("GGML_HEXAGON_ARCH");
+    const char * str_verbose  = getenv("GGML_HEXAGON_VERBOSE");
+    const char * str_hostbuf  = getenv("GGML_HEXAGON_HOSTBUF");
+    const char * str_opstage  = getenv("GGML_HEXAGON_OPSTAGE");
+    const char * str_opbatch  = getenv("GGML_HEXAGON_OPBATCH");
+    const char * str_opqueue  = getenv("GGML_HEXAGON_OPQUEUE");
+    const char * str_opfilter = getenv("GGML_HEXAGON_OPFILTER");
+    const char * str_profile  = getenv("GGML_HEXAGON_PROFILE");
+    const char * str_etm      = getenv("GGML_HEXAGON_ETM");
+    const char * str_nhvx     = getenv("GGML_HEXAGON_NHVX");
+    const char * str_use_hmx  = getenv("GGML_HEXAGON_USE_HMX");
+    const char * str_ndev     = getenv("GGML_HEXAGON_NDEV");
+    const char * str_arch     = getenv("GGML_HEXAGON_ARCH");
+    const char * str_vmem     = getenv("GGML_HEXAGON_VMEM");
+    const char * str_mbuf     = getenv("GGML_HEXAGON_MBUF");
+
+    // Init Arch first since it affects other defaults
+    if (!str_arch) {
+        int err = get_hex_arch_ver(CDSP_DOMAIN_ID, &opt_arch);
+        if (err != 0) {
+            GGML_LOG_ERROR("ggml-hex: failed to query HTP version (err %d) defaulting to v73\n", err);
+            opt_arch = 73;
+        }
+    } else {
+        if (str_arch[0] == 'v' || str_arch[0] == 'V') {
+            str_arch++;
+        }
+        opt_arch = strtoul(str_arch, NULL, 0);
+    }
+
+    size_t MiB = 1024 * 1024;
+
+    // Update vmem default
+    opt_vmem = opt_arch >= 75 ? HTP_OP_MAX_VMEM_DEFAULT : 3000 * MiB;
 
     auto RE_ICASE = std::regex_constants::icase;
 
-    opt_opfilter     = str_opfilter ? new std::regex(str_opfilter, RE_ICASE) : NULL;
-    opt_verbose      = str_verbose  ? atoi(str_verbose)             : 0;
-    opt_hostbuf      = str_hostbuf  ? atoi(str_hostbuf)             : opt_hostbuf;
-    opt_opstage      = str_opstage  ? strtoul(str_opstage, NULL, 0) : opt_opstage;
-    opt_opbatch      = str_opbatch  ? strtoul(str_opbatch, NULL, 0) : opt_opbatch;
-    opt_opqueue      = str_opqueue  ? strtoul(str_opqueue, NULL, 0) : opt_opqueue;
-    opt_etm          = str_etm      ? atoi(str_etm)                 : 0;
-    opt_nhvx         = str_nhvx     ? strtoul(str_nhvx, NULL, 0)    : opt_nhvx;
-    opt_use_hmx      = str_use_hmx  ? atoi(str_use_hmx)             : opt_use_hmx;
-    opt_ndev         = str_ndev     ? strtoul(str_ndev, NULL, 0)    : opt_ndev;
-    opt_hostbuf      = str_hostbuf  ? atoi(str_hostbuf)             : opt_hostbuf;
+    opt_opfilter  = str_opfilter ? new std::regex(str_opfilter, RE_ICASE) : NULL;
+    opt_verbose   = str_verbose  ? atoi(str_verbose)                      : 0;
+    opt_hostbuf   = str_hostbuf  ? atoi(str_hostbuf)                      : opt_hostbuf;
+    opt_opstage   = str_opstage  ? strtoul(str_opstage, NULL, 0)          : opt_opstage;
+    opt_opbatch   = str_opbatch  ? strtoul(str_opbatch, NULL, 0)          : opt_opbatch;
+    opt_opqueue   = str_opqueue  ? strtoul(str_opqueue, NULL, 0)          : opt_opqueue;
+    opt_profile   = str_profile  ? atoi(str_profile)                      : 0;
+    opt_etm       = str_etm      ? atoi(str_etm)                          : 0;
+    opt_nhvx      = str_nhvx     ? strtoul(str_nhvx, NULL, 0)             : opt_nhvx;
+    opt_use_hmx   = str_use_hmx  ? atoi(str_use_hmx)                      : opt_use_hmx;
+    opt_ndev      = str_ndev     ? strtoul(str_ndev, NULL, 0)             : opt_ndev;
+    opt_hostbuf   = str_hostbuf  ? atoi(str_hostbuf)                      : opt_hostbuf;
+    opt_mbuf      = str_mbuf     ? strtoul(str_mbuf, NULL, 0) * MiB       : opt_mbuf;
+    opt_vmem      = str_vmem     ? strtoul(str_vmem, NULL, 0) * MiB       : opt_vmem;
+
+    if (opt_ndev > GGML_HEXAGON_MAX_SESSIONS) {
+        opt_ndev = GGML_HEXAGON_MAX_SESSIONS;
+    }
+
+#if defined(__ANDROID__)
+    if (opt_arch < 75) {
+        opt_ndev = 1;
+        GGML_LOG_WARN("ggml-hex: forcing ndev to 1 for SoCs archs lower than v75.\n");
+    }
+#endif
 
     if (str_profile) {
         opt_pmu_evt = [&]() -> std::vector<uint32_t> {
@@ -3520,17 +3578,6 @@ static void ggml_hexagon_init(ggml_backend_reg * reg) {
                 vec_to_str<uint32_t, 16>(opt_pmu_evt).c_str());
     }
 
-    if (opt_ndev > GGML_HEXAGON_MAX_SESSIONS) {
-        opt_ndev = GGML_HEXAGON_MAX_SESSIONS;
-    }
-
-    if (str_arch) {
-        if (str_arch[0] == 'v') {
-            str_arch++;
-        }
-        opt_arch = strtoul(str_arch, NULL, 0);
-    }
-
     reg->context = new ggml_hexagon_registry(reg);
 }
 

ggml/src/ggml-hexagon/htp/CMakeLists.txt

@@ -44,6 +44,11 @@ target_compile_definitions(${HTP_LIB} PRIVATE
     $<IF:$<BOOL:${HEXAGON_HTP_DEBUG}>,FARF_HIGH=1,>
     FP32_QUANTIZE_GROUP_SIZE=${GGML_HEXAGON_FP32_QUANTIZE_GROUP_SIZE})
 
+if (GGML_HEXAGON_FA_EXP2_HF)
+    message(STATUS "ggml-htp: HMX_FA_USE_EXP2_HF=1 (use FP16 exp2 polynomial in FA softmax)")
+    target_compile_definitions(${HTP_LIB} PRIVATE HMX_FA_USE_EXP2_HF=1)
+endif()
+
 # HMX acceleration: available on v73+ architectures
 set(HTP_HMX_VERSIONS v73 v75 v79 v81)
 list(FIND HTP_HMX_VERSIONS ${DSP_VERSION} _hmx_idx)
@@ -52,11 +57,13 @@ if (_hmx_idx GREATER_EQUAL 0)
     target_sources(${HTP_LIB} PRIVATE
         hmx-queue.c
         hmx-matmul-ops.c
+        hmx-flash-attn-ops.c
     )
 
     # -mhmx enables HMX instruction set (needed by files that include hmx-utils.h)
     set_source_files_properties(
         hmx-matmul-ops.c
+        hmx-flash-attn-ops.c
         PROPERTIES COMPILE_OPTIONS "-mhmx"
     )
 

ggml/src/ggml-hexagon/htp/cmake-toolchain.cmake

@@ -138,15 +138,15 @@ set(CMAKE_SHARED_LIBRARY_SONAME_C_FLAG   "-Wl,-soname,")
 set(CMAKE_SHARED_LIBRARY_SONAME_CXX_FLAG "-Wl,-soname,")
 
 #Compiler Options
-set(COMMON_FLAGS "-mcpu=hexagon${V_ARCH} -m${V_ARCH} -mhvx=${V_ARCH} -fvectorize -Wall -Werror -fno-zero-initialized-in-bss -G0 -fdata-sections -fpic ${XQF_ARGS}")
+set(COMMON_FLAGS "-mcpu=hexagon${V_ARCH} -m${V_ARCH} -mhvx=${V_ARCH} -fvectorize -flto -Wall -Werror -fno-zero-initialized-in-bss -G0 -fdata-sections -fpic ${XQF_ARGS}")
 
 set(CMAKE_CXX_FLAGS_DEBUG          "${COMMON_FLAGS} -O0 -D_DEBUG -g")
-set(CMAKE_CXX_FLAGS_RELWITHDEBINFO "${COMMON_FLAGS} -O3 -g")
-set(CMAKE_CXX_FLAGS_RELEASE        "${COMMON_FLAGS} -O3")
+set(CMAKE_CXX_FLAGS_RELWITHDEBINFO "${COMMON_FLAGS} -O2 -g")
+set(CMAKE_CXX_FLAGS_RELEASE        "${COMMON_FLAGS} -O2")
 
 set(CMAKE_C_FLAGS_DEBUG            "${COMMON_FLAGS} -O0 -D_DEBUG -g")
-set(CMAKE_C_FLAGS_RELWITHDEBINFO   "${COMMON_FLAGS} -O3 -g")
-set(CMAKE_C_FLAGS_RELEASE          "${COMMON_FLAGS} -O3")
+set(CMAKE_C_FLAGS_RELWITHDEBINFO   "${COMMON_FLAGS} -O2 -g")
+set(CMAKE_C_FLAGS_RELEASE          "${COMMON_FLAGS} -O2")
 
 set(CMAKE_ASM_FLAGS_DEBUG          "${COMMON_FLAGS} ${CMAKE_CXX_FLAGS_DEBUG}")
 set(CMAKE_ASM_FLAGS_RELEASE        "${COMMON_FLAGS} ${CMAKE_CXX_FLAGS_RELEASE}")

ggml/src/ggml-hexagon/htp/flash-attn-ops.c

@@ -17,13 +17,14 @@
 #include "htp-ctx.h"
 #include "htp-ops.h"
 #include "htp-ops.h"
+#include "hmx-ops.h"
 
 // Must be multiple of 32
 #define FLASH_ATTN_BLOCK_SIZE (32 * 2)
 
 // This is a bit of a hack because the compiler is strugling to properly inline
 // the default hvx_vec_f32_to_f16 with output into the local array.
-static void __attribute__((noinline)) hvx_vec_f32_to_f16_a(void *ptr, HVX_Vector v0, HVX_Vector v1)
+static __attribute__((noinline)) void hvx_vec_f32_to_f16_a(void *ptr, HVX_Vector v0, HVX_Vector v1)
 {
     *(HVX_Vector *) ptr = hvx_vec_f32_to_f16(v0, v1);
 }
@@ -621,6 +622,17 @@ int op_flash_attn_ext(struct htp_ops_context * octx) {
         return HTP_STATUS_NO_SUPPORT;
     }
 
+#ifdef HTP_HAS_HMX
+    // HMX path: prefill (neq1 >= 32), head_dim multiple of 32, F16 KV
+    if (k->type == HTP_TYPE_F16 && v->type == HTP_TYPE_F16 && k->ne[0] % 32 == 0 && q->ne[1] >= 32) {
+        int ret = hmx_flash_attn_ext(octx);
+        if (ret == HTP_STATUS_OK) {
+            return ret;
+        }
+        // VTCM too small or other failure -> fall through to HVX path
+    }
+#endif
+
     struct htp_fa_context factx;
     factx.octx = octx;
 

ggml/src/ggml-hexagon/htp/hex-utils.h

@@ -74,6 +74,12 @@ static inline size_t hex_smax(size_t a, size_t b) {
     return a > b ? a : b;
 }
 
+static inline void hex_swap_ptr(void ** p1, void ** p2) {
+    void * t = *p1;
+    *p1      = *p2;
+    *p2      = t;
+}
+
 static inline void hex_l2fetch(const void * p, uint32_t width, uint32_t stride, uint32_t height) {
     const uint64_t control = Q6_P_combine_RR(stride, Q6_R_combine_RlRl(width, height));
     Q6_l2fetch_AP((void *) p, control);

ggml/src/ggml-hexagon/htp/hmx-ops.h

@@ -61,6 +61,9 @@ int hmx_mat_mul_permuted_qk_0_d16a32(struct htp_context *ctx,
                                       int m, int k, int n,
                                       int weight_type);
 
+// HMX flash attention
+int hmx_flash_attn_ext(struct htp_ops_context * octx);
+
 #ifdef __cplusplus
 }
 #endif

ggml/src/ggml-hexagon/htp/hmx-utils.h

@@ -4,6 +4,9 @@
 #ifndef HMX_UTILS_H
 #define HMX_UTILS_H
 
+#include "hvx-base.h"
+
+#include <assert.h>
 #include <hexagon_types.h>
 #include <stddef.h>
 
@@ -12,21 +15,188 @@
 #define HMX_FP16_TILE_N_ELMS 1024
 #define HMX_FP16_TILE_SIZE   2048
 
-#define HMX_INLINE_ALWAYS inline __attribute__((unused, always_inline))
-
 // Initialise aligned 256-byte area with scale vector + zero padding.
-static HMX_INLINE_ALWAYS void hmx_init_column_scales(void *out_scales, HVX_Vector v_scale) {
-    HVX_Vector *pv = (HVX_Vector *)out_scales;
-    *pv++ = v_scale;
-    *pv   = Q6_V_vzero();
+static inline void hmx_init_column_scales(void *out_scales, HVX_Vector v_scale) {
+    volatile HVX_Vector *pv = (HVX_Vector *) out_scales;
+    pv[0] = v_scale;
+    pv[1] = Q6_V_vzero();
 }
 
-// --- VTCM sequential allocator (from htp-ops-lib/include/dsp/vtcm_mgr.h) ---
+// --- Shared scatter offsets and interleave helper ---
 
-static inline uint8_t *vtcm_seq_alloc(uint8_t **vtcm_ptr, size_t size) {
-    uint8_t *p = *vtcm_ptr;
-    *vtcm_ptr += size;
-    return p;
+// vscatter offsets for fused dequant+transpose: write K-values directly to [K][N] tile.
+// word[i] = i*128 maps K-row-pair i to byte offset i*128.
+// Column offset (n*4) is added at runtime.  Entries 0..15 cover one tile (region 2047);
+// entries 16..31 cover the next adjacent tile (region 4095) — pick region size at the
+// call site to scatter into one tile (masked) or two contiguous tiles (unmasked).
+static const int32_t hmx_transpose_scatter_offsets[32] __attribute__((aligned(VLEN))) = {
+    0 * 128,  1 * 128,  2 * 128,  3 * 128,  4 * 128,  5 * 128,  6 * 128,  7 * 128,  8 * 128,  9 * 128,  10 * 128,
+    11 * 128, 12 * 128, 13 * 128, 14 * 128, 15 * 128, 16 * 128, 17 * 128, 18 * 128, 19 * 128, 20 * 128, 21 * 128,
+    22 * 128, 23 * 128, 24 * 128, 25 * 128, 26 * 128, 27 * 128, 28 * 128, 29 * 128, 30 * 128, 31 * 128,
+};
+
+// Scatter row-major FP16 data (in VTCM scratch) into transposed [K][N] tiles.
+// vtcm_src: [n_cols][src_stride] row-major fp16 (only first k elements per row are used)
+// vtcm_dst: [n_col_tiles][n_k_tiles][HMX_FP16_TILE_N_ELMS] tile-major interleaved fp16
+// Processes rows [start_row, end_row) for multi-thread slicing.
+// Full range: start_row=0, end_row=n_cols.
+static inline void hmx_interleave_rows_to_tiles(__fp16 * restrict vtcm_dst,
+                                            const __fp16 * restrict vtcm_src,
+                                            int n_cols,
+                                            int k,
+                                            int src_stride,
+                                            int start_row,
+                                            int end_row) {
+    assert(k % HMX_FP16_TILE_N_COLS == 0);
+
+    const int            n_k_tiles     = k / HMX_FP16_TILE_N_COLS;
+    const HVX_Vector     v_scat_base   = hvx_vmem(hmx_transpose_scatter_offsets);
+    const HVX_Vector     v_scat_step   = Q6_V_vsplat_R(4);
+    const HVX_VectorPred q_mask64      = Q6_Q_vsetq_R(64);
+    // Each hvx_vmemu load brings 64 fp16 = 128 bytes covering 2 adjacent K-tiles.
+    // When n_k_tiles is even, scatter into 2 K-tiles per call (region 4095, no mask)
+    // using the upper half of hmx_transpose_scatter_offsets.  Tail one K-tile (when
+    // n_k_tiles is odd) falls back to single-tile masked scatter.
+    const bool           pair_scatter  = (n_k_tiles & 1) == 0;
+    const size_t         pair_region   = (size_t) (2 * HMX_FP16_TILE_SIZE - 1);
+    const size_t         single_region = (size_t) (HMX_FP16_TILE_SIZE - 1);
+    __builtin_assume(k > 0);
+    __builtin_assume(end_row > start_row);
+
+    if (pair_scatter) {
+        // Step c by 64 fp16 (two K-tiles per scatter), advance dst by 2 tiles per iter.
+        const int    c_step      = 2 * HMX_FP16_TILE_N_COLS;
+        const size_t c_byte_step = (size_t) c_step * sizeof(__fp16);
+        const size_t dst_step    = 2 * (size_t) HMX_FP16_TILE_N_ELMS;
+        const int    n_c_iters   = k / c_step;
+
+        for (int r = start_row; r < end_row; r += 2) {
+            const int        ct             = r / HMX_FP16_TILE_N_ROWS;
+            const int        local_r        = r % HMX_FP16_TILE_N_ROWS;
+            const bool       next_row_valid = (r + 1) < end_row && (r + 1) < n_cols;
+            const HVX_Vector v_off0         = Q6_Vw_vadd_VwVw(v_scat_base, Q6_V_vsplat_R(local_r * 4));
+            const HVX_Vector v_off1         = Q6_Vw_vadd_VwVw(v_off0, v_scat_step);
+
+            __fp16 *        tile_base = vtcm_dst + (size_t) ct * n_k_tiles * HMX_FP16_TILE_N_ELMS;
+            const uint8_t * p0        = (const uint8_t *) (vtcm_src + r * src_stride);
+            const uint8_t * p1        = next_row_valid ? (const uint8_t *) (vtcm_src + (r + 1) * src_stride) : NULL;
+
+            if (p1) {
+                for (int i = 0; i < n_c_iters; ++i) {
+                    HVX_Vector v0 = hvx_vmemu(p0);
+                    p0 += c_byte_step;
+                    HVX_Vector v1 = hvx_vmemu(p1);
+                    p1 += c_byte_step;
+                    Q6_vscatter_RMVwV((size_t) tile_base, pair_region, v_off0, v0);
+                    Q6_vscatter_RMVwV((size_t) tile_base, pair_region, v_off1, v1);
+                    tile_base += dst_step;
+                }
+            } else {
+                const HVX_Vector vzero = Q6_V_vzero();
+                for (int i = 0; i < n_c_iters; ++i) {
+                    HVX_Vector v0 = hvx_vmemu(p0);
+                    p0 += c_byte_step;
+                    Q6_vscatter_RMVwV((size_t) tile_base, pair_region, v_off0, v0);
+                    Q6_vscatter_RMVwV((size_t) tile_base, pair_region, v_off1, vzero);
+                    tile_base += dst_step;
+                }
+            }
+        }
+    } else {
+        // Fallback: scatter one K-tile per call (region 2047, masked).
+        const int    c_step      = HMX_FP16_TILE_N_COLS;
+        const size_t c_byte_step = (size_t) c_step * sizeof(__fp16);
+        const size_t dst_step    = (size_t) HMX_FP16_TILE_N_ELMS;
+        const int    n_c_iters   = k / c_step;
+
+        for (int r = start_row; r < end_row; r += 2) {
+            const int        ct             = r / HMX_FP16_TILE_N_ROWS;
+            const int        local_r        = r % HMX_FP16_TILE_N_ROWS;
+            const bool       next_row_valid = (r + 1) < end_row && (r + 1) < n_cols;
+            const HVX_Vector v_off0         = Q6_Vw_vadd_VwVw(v_scat_base, Q6_V_vsplat_R(local_r * 4));
+            const HVX_Vector v_off1         = Q6_Vw_vadd_VwVw(v_off0, v_scat_step);
+
+            __fp16 *        tile_base = vtcm_dst + (size_t) ct * n_k_tiles * HMX_FP16_TILE_N_ELMS;
+            const uint8_t * p0        = (const uint8_t *) (vtcm_src + r * src_stride);
+            const uint8_t * p1        = next_row_valid ? (const uint8_t *) (vtcm_src + (r + 1) * src_stride) : NULL;
+
+            if (p1) {
+                for (int i = 0; i < n_c_iters; ++i) {
+                    HVX_Vector v0 = hvx_vmemu(p0);
+                    p0 += c_byte_step;
+                    HVX_Vector v1 = hvx_vmemu(p1);
+                    p1 += c_byte_step;
+                    Q6_vscatter_QRMVwV(q_mask64, (size_t) tile_base, single_region, v_off0, v0);
+                    Q6_vscatter_QRMVwV(q_mask64, (size_t) tile_base, single_region, v_off1, v1);
+                    tile_base += dst_step;
+                }
+            } else {
+                const HVX_Vector vzero = Q6_V_vzero();
+                for (int i = 0; i < n_c_iters; ++i) {
+                    HVX_Vector v0 = hvx_vmemu(p0);
+                    p0 += c_byte_step;
+                    Q6_vscatter_QRMVwV(q_mask64, (size_t) tile_base, single_region, v_off0, v0);
+                    Q6_vscatter_QRMVwV(q_mask64, (size_t) tile_base, single_region, v_off1, vzero);
+                    tile_base += dst_step;
+                }
+            }
+        }
+    }
+}
+
+// Interleave row-major FP16 data into column-major tile format.
+// Input: [n_rows, head_dim] row-major.  Output: tile[dim_tile][row_tile].
+// Processes rows [start_row, end_row) for multi-thread slicing.
+// Full range: start_row=0, end_row=n_rows.
+static inline void hmx_interleave_cols_to_tiles(__fp16 * restrict tiles_out,
+                                            const __fp16 * restrict src,
+                                            int n_rows,
+                                            int head_dim,
+                                            int src_stride,
+                                            int n_row_tiles,
+                                            int start_row,
+                                            int end_row) {
+    __builtin_assume(head_dim > 0);
+    const size_t tile_stride_elms = (size_t) n_row_tiles * HMX_FP16_TILE_N_ELMS;
+
+    for (int r = start_row; r < end_row; r += 2) {
+        const bool next_row_valid = (r + 1) < end_row && (r + 1) < n_rows;
+
+        const HVX_Vector * pv_in0 = (const HVX_Vector *) (src + r * src_stride);
+        const HVX_Vector * pv_in1 = next_row_valid ? (const HVX_Vector *) (src + (r + 1) * src_stride) : NULL;
+
+        // Row-pair invariants hoisted out of the c loop.
+        const int r0      = r / HMX_FP16_TILE_N_ROWS;
+        const int r1_half = (r % HMX_FP16_TILE_N_ROWS) / 2;
+
+        // tb0 starts at tile (c0=0, r0); tb1 at the adjacent dim-tile (c0=1, r0).
+        // Each c step (+= 64) advances both by 2 dim-tiles worth of fp16.
+        __fp16 *     tb0     = tiles_out + (size_t) r0 * HMX_FP16_TILE_N_ELMS;
+        __fp16 *     tb1     = tb0 + tile_stride_elms;
+        const size_t tb_step = 2 * tile_stride_elms;
+
+        if (pv_in1) {
+            for (int c = 0; c < head_dim; c += 64) {
+                HVX_Vector     v0             = *pv_in0++;
+                HVX_Vector     v1             = *pv_in1++;
+                HVX_VectorPair vp             = Q6_W_vshuff_VVR(v1, v0, -2);
+                ((HVX_Vector *) tb0)[r1_half] = Q6_V_lo_W(vp);
+                ((HVX_Vector *) tb1)[r1_half] = Q6_V_hi_W(vp);
+                tb0 += tb_step;
+                tb1 += tb_step;
+            }
+        } else {
+            const HVX_Vector vzero = Q6_V_vzero();
+            for (int c = 0; c < head_dim; c += 64) {
+                HVX_Vector     v0             = *pv_in0++;
+                HVX_VectorPair vp             = Q6_W_vshuff_VVR(vzero, v0, -2);
+                ((HVX_Vector *) tb0)[r1_half] = Q6_V_lo_W(vp);
+                ((HVX_Vector *) tb1)[r1_half] = Q6_V_hi_W(vp);
+                tb0 += tb_step;
+                tb1 += tb_step;
+            }
+        }
+    }
 }
 
 #endif // HMX_UTILS_H

ggml/src/ggml-hexagon/htp/htp-ctx.h

@@ -20,7 +20,7 @@ struct htp_mmap {
     uint64_t size;
     uint64_t base;
     uint32_t fd;
-    uint32_t pinned;
+    uint32_t reserved;
 };
 
 // Scratchpad state
@@ -77,6 +77,8 @@ struct htp_context {
     atomic_bool            vtcm_valid;
     atomic_bool            vtcm_needs_release;
 
+    uint64_t               max_vmem;
+
     struct htp_ops_context octx;
 
 #ifdef HTP_HAS_HMX

ggml/src/ggml-hexagon/htp/htp-ops.h

@@ -90,15 +90,11 @@ enum htp_op_code {
 #define HTP_OP_MAX_INPUTS  6    // aka GGML_MAX_SRCS
 #define HTP_OP_MAX_PARAMS  16   // aka GGML_MAX_OP_PARAMS
 
-#define HTP_OP_MAX_BUFS    8
+#define HTP_OP_MAX_BUFS    16
 #define HTP_OP_MAX_REQS    256
 #define HTP_OP_MAX_TENSORS (HTP_OP_MAX_REQS * HTP_OP_MAX_INPUTS + HTP_OP_MAX_REQS)
 
-#if __HVX_ARCH__ < 75
-#define HTP_OP_MAX_VMEM    (3167538380u)
-#else
-#define HTP_OP_MAX_VMEM    (3221225472u)
-#endif
+#define HTP_OP_MAX_VMEM_DEFAULT (3355443200u)
 
 #define HTP_MMAP_MAX_VMEM  (2147483648u)