not there yet

kliuae · kliuae · commit 2f056b974344 · 2024-12-20T14:49:08.000Z
diff --git a/vllm/config.py b/vllm/config.py
@@ -427,7 +427,7 @@ def _verify_quantization(self) -> None:
         optimized_quantization_methods = [
             "fp8", "marlin", "modelopt", "gptq_marlin_24", "gptq_marlin",
             "awq_marlin", "fbgemm_fp8", "compressed_tensors",
-            "compressed-tensors", "experts_int8", "ptpc_fp8"
+            "compressed-tensors", "experts_int8"
         ]
         if self.quantization is not None:
             self.quantization = self.quantization.lower()

Original file line number	Diff line number	Diff line change
`@@ -427,7 +427,7 @@ def _verify_quantization(self) -> None:`
`427`	`427`	`optimized_quantization_methods = [`
`428`	`428`	`"fp8", "marlin", "modelopt", "gptq_marlin_24", "gptq_marlin",`
`429`	`429`	`"awq_marlin", "fbgemm_fp8", "compressed_tensors",`
`430`		`- "compressed-tensors", "experts_int8", "ptpc_fp8"`
	`430`	`+ "compressed-tensors", "experts_int8"`
`431`	`431`	`]`
`432`	`432`	`if self.quantization is not None:`
`433`	`433`	`self.quantization = self.quantization.lower()`