[CPU][Llama4] Fix Llama4 MoE inputs with "apply_router_weight_on_input" (#7889)

2025-07-18 12:43:25 +08:00
parent 8aa5ae6b04
commit 48c1fa7bb6
5 changed files with 35 additions and 4 deletions
--- a/python/sglang/srt/configs/update_config.py
+++ b/python/sglang/srt/configs/update_config.py
@@ -115,5 +115,7 @@ def adjust_config_with_unaligned_cpu_tp(
    model_config = update_intermediate_size(
        model_config, "intermediate_size", intermediate_padding_size
    )
-
+    model_config = update_intermediate_size(
+        model_config, "intermediate_size_mlp", intermediate_padding_size
+    )
    return model_config