MCPcopy Index your code
hub / github.com/InternLM/lmdeploy / model_format

Method model_format

lmdeploy/cli/utils.py:151–160  ·  view source on GitHub ↗
(parser, default: str = None)

Source from the content-addressed store, hash-verified

149
150 @staticmethod
151 def model_format(parser, default: str = None):
152 return parser.add_argument('--model-format',
153 type=str,
154 default=default,
155 choices=['hf', 'awq', 'gptq', 'compressed-tensors', 'fp8', 'mxfp4'],
156 help='The format of input model. `hf` means `hf_llama`, '
157 '`awq` and `gptq` refer to 4-bit grouped quantization, '
158 '`compressed-tensors` refers to pack-quantized grouped int4 checkpoints and is '
159 'usually auto-detected from the model config, `fp8` refers to blocked fp8 '
160 'checkpoints, and `mxfp4` refers to MXFP4 expert weights.')
161
162 @staticmethod
163 def revision(parser, default: str = None):

Callers 5

add_parser_chatMethod · 0.80
add_parser_api_serverMethod · 0.80
parse_argsFunction · 0.80
parse_argsFunction · 0.80
parse_argsFunction · 0.80

Calls

no outgoing calls

Tested by

no test coverage detected