[Bugfix] GPT OSS Attritbute error on H100 (#25228)

Signed-off-by: Varun Sundar Rabindranath <vsundarr@redhat.com> Co-authored-by: Varun Sundar Rabindranath <vsundarr@redhat.com>
2025-09-19 15:14:09 -04:00
parent a2a5f79e09
commit 7852b82b93
2 changed files with 4 additions and 4 deletions
--- a/vllm/model_executor/layers/quantization/mxfp4.py
+++ b/vllm/model_executor/layers/quantization/mxfp4.py
@@ -638,8 +638,8 @@ class Mxfp4MoEMethod(FusedMoEMethodBase):
            return None

        if self.mxfp4_backend == Mxfp4Backend.TRITON:
-            w1_scale = layer.w13_precision_config
-            w2_scale = layer.w2_precision_config
+            w1_scale = self.w13_precision_config
+            w2_scale = self.w2_precision_config
        else:
            w1_scale = layer.w13_weight_scale
            w2_scale = layer.w2_weight_scale