[Bugfix] Fix null modules_to_not_convert in FBGEMM Fp8 quantization (#6665)

2024-07-22 19:25:05 -07:00 · 2024-07-22 19:25:05 -07:00 · c5e8330997
commit c5e8330997
parent e0c15758b8
1 changed files with 1 additions and 1 deletions
--- a/vllm/model_executor/layers/quantization/fbgemm_fp8.py
+++ b/vllm/model_executor/layers/quantization/fbgemm_fp8.py
@ -31,7 +31,7 @@ class FBGEMMFp8Config(QuantizationConfig):
    """Config class for FBGEMM Fp8."""

    def __init__(self, ignore_list: List[str], input_scale_ub: float):
-        self.ignore_list = ignore_list
+        self.ignore_list = ignore_list if ignore_list else []
        self.input_scale_ub = input_scale_ub

        # For GPUs that lack FP8 hardware support, we can leverage the Marlin