nwpie/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2026-02-05 13:53:23 +02:00
Code Issues Packages Projects Releases Wiki Activity

models : fix backend assignment for Granite/Nemotron graphs (#18599)

Browse Source 
* models : fix backend assignment for Granite/Nemotron graphs

* cont : add ref

* cont : move call to build_inp_embd()
This commit is contained in:
Georgi Gerganov
2026-01-05 12:34:23 +02:00
committed by GitHub
parent b37124d2d2
commit 2da64a2f8a
1 changed files with 4 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
src/llama-graph.cpp
View File

@@ -1326,6 +1326,10 @@ ggml_tensor * llm_graph_context::build_inp_embd(ggml_tensor * tok_embd) const {
res->add_input(std::move(inp));
// make sure the produced embeddings are immediately materialized in the ggml graph
// ref: https://github.com/ggml-org/llama.cpp/pull/18599
ggml_build_forward_expand(gf, cur);
return cur;
}