[Quant] Add SupportsQuant to phi3 and clip (#13104)

2025-02-15 22:28:33 -05:00
parent 80f63a3966
commit 12913d17ba
25 changed files with 67 additions and 13 deletions
--- a/vllm/model_executor/layers/quantization/bitsandbytes.py
+++ b/vllm/model_executor/layers/quantization/bitsandbytes.py
@@ -30,7 +30,7 @@ class BitsAndBytesConfig(QuantizationConfig):
        llm_int8_skip_modules: Optional[List[str]] = None,
        llm_int8_threshold: float = 6.0,
    ) -> None:
-
+        super().__init__()
        self.load_in_8bit = load_in_8bit
        self.load_in_4bit = load_in_4bit
        self.bnb_4bit_compute_dtype = bnb_4bit_compute_dtype