diff --git a/python/sglang/srt/layers/quantization/mxfp4.py b/python/sglang/srt/layers/quantization/mxfp4.py
index 4616879ce..84424458c 100644
--- a/python/sglang/srt/layers/quantization/mxfp4.py
+++ b/python/sglang/srt/layers/quantization/mxfp4.py
@@ -866,5 +866,6 @@ class Mxfp4DynamicQuantMoEMethod(FusedMoEMethodBase):
                 else ActivationType.Gelu
             ),
             doweight_stage1=False,
+            expert_mask=layer.expert_mask_gpu,
         )
         return StandardCombineInput(hidden_states=output)
diff --git a/python/sglang/srt/layers/quantization/quark/quark_moe.py b/python/sglang/srt/layers/quantization/quark/quark_moe.py
index e48392201..186f6c5d0 100644
--- a/python/sglang/srt/layers/quantization/quark/quark_moe.py
+++ b/python/sglang/srt/layers/quantization/quark/quark_moe.py
@@ -245,6 +245,7 @@ class QuarkW4A4MXFp4MoEMethod(QuarkMoEMethod):
                 else ActivationType.Gelu
             ),
             doweight_stage1=False,
+            expert_mask=layer.expert_mask_gpu,
         )
         return StandardCombineInput(hidden_states=output)