EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Tiny refactor ModelConfig.from_server_args (#5219)

Browse Source
This commit is contained in:
fzyzcjy
2025-05-08 16:02:43 +08:00
committed by GitHub
parent 3b2680a44d
commit b6cf3532b5
6 changed files with 23 additions and 53 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
test/srt/test_gptqmodel_dynamic.py
View File

@@ -43,16 +43,7 @@ def check_quant_method(model_path: str, use_marlin_kernel: bool):
pass
server_args = ServerArgs(model_path=model_path, dtype=torch.float16)
model_config = ModelConfig(
server_args.model_path,
trust_remote_code=server_args.trust_remote_code,
revision=server_args.revision,
context_length=server_args.context_length,
model_override_args=server_args.json_model_override_args,
is_embedding=server_args.is_embedding,
dtype=server_args.dtype,
quantization=server_args.quantization,
)
model_config = ModelConfig.from_server_args(server_args)
load_config = LoadConfig()
device_config = DeviceConfig("cuda")