model: apply .scale tensors automatically + fix correctness (#32669)

* model: apply .scale tensors automatically

* rm build_moe_ffn "_s" overload

* use llama_prec_policy

* fix ci

* build_lora_mm_id: rm w_s
This commit is contained in:
Xuan-Son Nguyen authored and GitHub committed 2026-11-21 16:30:01 +01:00
1 parent e50c254a22
commit 19bc41dbb4
39 files changed
+164 -290

No files matched your search

-3
View File
@@ -277,9 +277,6 @@ ggml_tensor * llama_model_zaya1::graph::build_router_moe(const llama_model & mod
il,
logits,
layer.ffn_gate_up_exps,
nullptr,
nullptr,
nullptr,
expert);
return ggml_mul(ctx0, moe, keep);