vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

ggml-cpu: Add CPU backend support for KleidiAI library

Browse source
This commit is contained in:
Charles Xu 2025-01-24 10:17:04 +01:00
parent c07e87f38b
commit 6adca19c94
19 changed files with 675 additions and 22 deletions
Download patch file Download diff file Expand all files Collapse all files

2
common/common.cpp
View file

@ -1099,6 +1099,8 @@ struct llama_model_params common_model_params_to_llama(common_params & params) {
mparams.kv_overrides = params.kv_overrides.data();
}
mparams.n_threads = params.cpuparams.n_threads;
return mparams;
}