sleepy/llama.cpp
1
0
Fork 0
Code Issues 16 Pull Requests Actions 30 Packages Projects Releases Wiki Activity

model : detect GigaChat3-10-A1.8B as deepseek lite (#17420)

Browse Source 
* Detect GigaChat3-10-A1.8B as deepseek lite

Hardcodes checking number of layers to detect if lite version of deepseek.

* Add commnent identifying deepseek lite variants

deepseek lite variants include DeepSeek-V2-Lite, GigaChat3-10B-A1.8B
This commit is contained in:
ubergarm
2025-11-21 08:51:38 -05:00
committed by GitHub
parent 28175f857d
commit 23bc779a6e
2 changed files with 6 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/models/deepseek2.cpp
+2 -1
View File
@@ -4,7 +4,8 @@
llm_build_deepseek2::llm_build_deepseek2(const llama_model & model, const llm_graph_params & params) :
llm_graph_context(params) {
bool is_lite = (hparams.n_layer == 27);
// lite variants include DeepSeek-V2-Lite, GigaChat3-10B-A1.8B
bool is_lite = (hparams.n_layer == 27 || hparams.n_layer == 26);
const bool is_mla = (hparams.n_embd_head_k_mla != 0 && hparams.n_embd_head_v_mla != 0);