biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Misc][ModelScope] Change to use runtime VLLM_USE_MODELSCOPE (#18655)

Browse Source 
Signed-off-by: Mengqing Cao <cmq0113@163.com>
Signed-off-by: Isotr0py <2037008807@qq.com>
Co-authored-by: Isotr0py <2037008807@qq.com>
This commit is contained in:
Mengqing Cao
2025-05-25 12:51:21 +08:00
committed by GitHub
parent cebc22f3b6
commit 6ab681bcbe
5 changed files with 19 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
vllm/model_executor/model_loader/default_loader.py
View File

@@ -11,8 +11,8 @@ import torch
from torch import nn
from transformers.utils import SAFE_WEIGHTS_INDEX_NAME
from vllm import envs
from vllm.config import LoadConfig, LoadFormat, ModelConfig, VllmConfig
from vllm.envs import VLLM_USE_MODELSCOPE
from vllm.logger import init_logger
from vllm.model_executor.model_loader.base_loader import BaseModelLoader
from vllm.model_executor.model_loader.utils import (
@@ -64,7 +64,7 @@ class DefaultModelLoader(BaseModelLoader):
Returns the path to the downloaded model, or None if the model is not
downloaded from ModelScope."""
if VLLM_USE_MODELSCOPE:
if envs.VLLM_USE_MODELSCOPE:
# download model from ModelScope hub,
# lazy import so that modelscope is not required for normal use.
# pylint: disable=C.