[Minor] Rename quantization nvfp4 to modelopt_fp4 (#18356)
Signed-off-by: mgoin <mgoin64@gmail.com>
This commit is contained in:
+1
-1
@@ -824,7 +824,7 @@ class ModelConfig:
|
||||
optimized_quantization_methods = [
|
||||
"fp8", "marlin", "modelopt", "gptq_marlin_24", "gptq_marlin",
|
||||
"awq_marlin", "fbgemm_fp8", "compressed-tensors", "experts_int8",
|
||||
"quark", "nvfp4", "bitblas", "gptq_bitblas"
|
||||
"quark", "modelopt_fp4", "bitblas", "gptq_bitblas"
|
||||
]
|
||||
if self.quantization is not None:
|
||||
self.quantization = cast(QuantizationMethods,
|
||||
|
||||
@@ -14,7 +14,7 @@ QuantizationMethods = Literal[
|
||||
"ptpc_fp8",
|
||||
"fbgemm_fp8",
|
||||
"modelopt",
|
||||
"nvfp4",
|
||||
"modelopt_fp4",
|
||||
"marlin",
|
||||
"bitblas",
|
||||
"gguf",
|
||||
@@ -120,7 +120,7 @@ def get_quantization_config(quantization: str) -> type[QuantizationConfig]:
|
||||
"fp8": Fp8Config,
|
||||
"fbgemm_fp8": FBGEMMFp8Config,
|
||||
"modelopt": ModelOptFp8Config,
|
||||
"nvfp4": ModelOptNvFp4Config,
|
||||
"modelopt_fp4": ModelOptNvFp4Config,
|
||||
"marlin": MarlinConfig,
|
||||
"bitblas": BitBLASConfig,
|
||||
"gguf": GGUFConfig,
|
||||
|
||||
@@ -192,7 +192,7 @@ class ModelOptNvFp4Config(QuantizationConfig):
|
||||
|
||||
@classmethod
|
||||
def get_name(cls) -> QuantizationMethods:
|
||||
return "nvfp4"
|
||||
return "modelopt_fp4"
|
||||
|
||||
@classmethod
|
||||
def get_supported_act_dtypes(cls) -> list[torch.dtype]:
|
||||
|
||||
Reference in New Issue
Block a user