model : support NVFP4 tensors for Gemma4 (#21971)

* support nvfp4 tensors for Gemma4 * add wo_s to build_attn * add wo_s to build_attn * fix glm4
2026-04-16 16:51:47 +02:00
parent b572d1ecd6
commit f772f6e434
105 changed files with 149 additions and 148 deletions
@@ -120,7 +120,7 @@ llm_build_plm::llm_build_plm(const llama_model & model, const llm_graph_params &
            cb(k_states, "k_states", il);

            cur = build_attn(inp_attn,
-                    model.layers[il].wo, NULL,
+                    model.layers[il].wo, NULL, model.layers[il].wo_s,
                    q_states, k_states, v_states, nullptr, nullptr, nullptr, kq_scale, il);
        }
        if (il == n_layer - 1 && inp_out_ids) {