MCPcopy
hub / github.com/lm-sys/FastChat / AWQConfig

Class AWQConfig

fastchat/modules/awq.py:10–21  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

8
9@dataclass
10class AWQConfig:
11 ckpt: str = field(
12 default=None,
13 metadata={
14 "help": "Load quantized model. The path to the local AWQ checkpoint."
15 },
16 )
17 wbits: int = field(default=16, metadata={"help": "#bits to use for quantization"})
18 groupsize: int = field(
19 default=-1,
20 metadata={"help": "Groupsize to use for quantization; default uses full row."},
21 )
22
23
24def load_awq_quantized(model_name, awq_config: AWQConfig, device):

Callers 2

create_model_workerFunction · 0.90
mainFunction · 0.90

Calls

no outgoing calls

Tested by

no test coverage detected

Used in the wild real call sites across dependent graphs

searching dependent graphs…