diff --git a/vllm/model_executor/layers/quantization/mxfp4.py b/vllm/model_executor/layers/quantization/mxfp4.py
index 96297c0c4d723..6823fa02a32d7 100644
--- a/vllm/model_executor/layers/quantization/mxfp4.py
+++ b/vllm/model_executor/layers/quantization/mxfp4.py
@@ -794,7 +794,8 @@ class Mxfp4MoEMethod(FusedMoEMethodBase):
                 )
             else:
                 raise NotImplementedError(
-                    "Incompatible Mxfp4 backend for EP batched experts format"
+                    f"Incompatible Mxfp4 backend ({self.mxfp4_backend}) for "
+                    "EP batched experts format"
                 )
         else:
             assert self.moe_quant_config is not None
@@ -813,8 +814,12 @@ class Mxfp4MoEMethod(FusedMoEMethodBase):
                 return TrtLlmGenExperts(self.moe, self.moe_quant_config, **kwargs)
             elif self.mxfp4_backend == Mxfp4Backend.MARLIN:
                 return MarlinExperts(self.moe_quant_config)
-            else:
+            elif self.mxfp4_backend == Mxfp4Backend.TRITON:
                 return OAITritonExperts(self.moe_quant_config)
+            else:
+                raise NotImplementedError(
+                    f"Incompatible Mxfp4 backend ({self.mxfp4_backend}) for EP"
+                )
 
     def _route_and_experts(
         self,