mikhail/llama.cpp-mtp-turboquant
1
0
Fork 0
Code Issues Pull Requests Actions 24 Packages Projects Releases Wiki Activity

model : support NVFP4 tensors for Gemma4 (#21971)

Browse Source 
* support nvfp4 tensors for Gemma4

* add wo_s to build_attn

* add wo_s to build_attn

* fix glm4
This commit is contained in:
Sigbjørn Skjæret
2026-04-16 16:51:47 +02:00
committed by GitHub
parent b572d1ecd6
commit f772f6e434
105 changed files with 149 additions and 148 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/models/cogvlm.cpp
+4 -2
View File
@@ -28,18 +28,20 @@ llm_build_cogvlm::llm_build_cogvlm(const llama_model & model, const llm_graph_pa
for (int il = 0; il < n_layer; ++il) {
// get either the text or image weight tensors
ggml_tensor *wqkv, *wo;
ggml_tensor *wqkv, *wo, *wo_s;
ggml_tensor *ffn_gate, *ffn_down, *ffn_up;
if (is_text) {
wqkv = model.layers[il].wqkv;
wo = model.layers[il].wo;
wo_s = model.layers[il].wo_s;
ffn_gate = model.layers[il].ffn_gate;
ffn_down = model.layers[il].ffn_down;
ffn_up = model.layers[il].ffn_up;
} else {
wqkv = model.layers[il].visexp_attn_wqkv;
wo = model.layers[il].visexp_attn_wo;
wo_s = nullptr;
ffn_gate = model.layers[il].visexp_ffn_gate;
ffn_down = model.layers[il].visexp_ffn_down;
ffn_up = model.layers[il].visexp_ffn_up;
@@ -64,7 +66,7 @@ llm_build_cogvlm::llm_build_cogvlm(const llama_model & model, const llm_graph_pa
Kcur = ggml_rope(ctx0, Kcur, inp_pos, n_embd_head, rope_type);
cur = build_attn(inp_attn,
wo, nullptr,
wo, nullptr, wo_s,
Qcur, Kcur, Vcur,
nullptr, nullptr, nullptr,
kq_scale, il);