happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

model : nit, DeepSeek V1 MoE is 16B and GigaChat is 20B (#12652) 

* nit, DeepSeek V1 MoE is 16B

* base type on n_ff_exp instead
This commit is contained in:
Sigbjørn Skjæret 2025-12-09 12:15:06 +01:00 committed by GitHub
parent 4e842d5120
commit 42b12b5608
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 3 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
src/llama-model.cpp
View File

@ -1606,8 +1606,9 @@ void llama_model::load_hparams(llama_model_loader & ml) {
ml.get_key(LLM_KV_EXPERT_SHARED_COUNT, hparams.n_expert_shared);
ml.get_key(LLM_KV_EXPERT_WEIGHTS_SCALE, hparams.expert_weights_scale);
switch (hparams.n_layer) {
case 28: type = LLM_TYPE_20B; break;
switch (hparams.n_ff_exp) {
case 1408: type = LLM_TYPE_16B; break;
case 1792: type = LLM_TYPE_20B; break;
default: type = LLM_TYPE_UNKNOWN;
}
} break;