Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

model : add NVFP4 MTP scale tensors (#23563)

Browse Source 
* Add NVFP4 MTP scale tensors

* Link Qwen3.5 MTP tensors

* Aligned nullptr
This commit is contained in:
Michael Wand
2026-05-23 13:30:31 +02:00
committed by GitHub
parent a497476330
commit b0df4c0cfd
4 changed files with 28 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/models/qwen35.cpp
+3 -2
View File
@@ -538,7 +538,7 @@ llama_model_qwen35::graph_mtp::graph_mtp(const llama_model & model, const llm_gr
ggml_tensor * concat = ggml_concat(ctx0, e_norm, h_norm, /*dim=*/ 0);
cb(concat, "mtp_concat", il);
ggml_tensor * cur = build_lora_mm(layer.nextn.eh_proj, concat);
ggml_tensor * cur = build_lora_mm(layer.nextn.eh_proj, concat, layer.nextn.eh_proj_s);
cb(cur, "mtp_eh_proj", il);
ggml_tensor * inpSA = cur;
@@ -626,8 +626,9 @@ llama_model_qwen35::graph_mtp::graph_mtp(const llama_model & model, const llm_gr
cb(cur, "mtp_shared_head_norm", -1);
ggml_tensor * head_w = layer.nextn.shared_head_head ? layer.nextn.shared_head_head : model.output;
ggml_tensor * head_s = layer.nextn.shared_head_head ? layer.nextn.shared_head_head_s : model.output_s;
GGML_ASSERT(head_w && "QWEN35 MTP: missing LM head (nextn.shared_head_head or model.output)");
cur = build_lora_mm(head_w, cur);
cur = build_lora_mm(head_w, cur, head_s);
cb(cur, "result_output", -1);
res->t_logits = cur;