[CI Failure] Fix torchao dep failure for Quantization Test (#26824)

Signed-off-by: mgoin <mgoin64@gmail.com>
2025-10-14 19:41:43 -04:00
parent 4aed506b65
commit 7e0ef4084a
4 changed files with 8 additions and 4 deletions
--- a/vllm/model_executor/layers/quantization/rtn.py
+++ b/vllm/model_executor/layers/quantization/rtn.py
@@ -15,6 +15,7 @@ from vllm.model_executor.layers.fused_moe.config import (
    FusedMoEConfig,
    FusedMoEQuantConfig,
 )
+from vllm.model_executor.layers.fused_moe.fused_marlin_moe import fused_marlin_moe
 from vllm.model_executor.layers.fused_moe.layer import FusedMoE, FusedMoEMethodBase
 from vllm.model_executor.layers.linear import (
    LinearBase,
@@ -396,7 +397,7 @@ class RTNMoEMethod(FusedMoEMethodBase):
            indices_type=self.topk_indices_dtype,
        )

-        return torch.ops.vllm.fused_marlin_moe(
+        return fused_marlin_moe(
            x,
            layer.w13_weight,
            layer.w2_weight,