[Bugfix] Fix misnamed parameter in compressed_tensors_moe.py (#35813)

Signed-off-by: Bill Nell <bnell@redhat.com> Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
2026-03-03 16:29:57 -05:00
parent e7213003cb
commit a9b8b13e5c
1 changed files with 1 additions and 1 deletions
--- a/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors_moe.py
+++ b/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors_moe.py
@@ -887,7 +887,7 @@ class CompressedTensorsW8A8Fp8MoEMethod(CompressedTensorsMoEMethod):
                w13,
                w13_scale,
                shard_size=layer.intermediate_size_per_partition,
-                num_experts=layer.num_local_experts,
+                num_experts=layer.local_num_experts,
                is_act_and_mul=self.moe.is_act_and_mul,
            )