[chore] Remove models weight on S3 logic (#27725)
Signed-off-by: kevin <kevin@anyscale.com>
This commit is contained in:
@@ -75,7 +75,6 @@ from vllm.platforms import CpuArchEnum, current_platform
|
||||
from vllm.plugins import load_general_plugins
|
||||
from vllm.ray.lazy_utils import is_in_ray_actor, is_ray_initialized
|
||||
from vllm.reasoning import ReasoningParserManager
|
||||
from vllm.test_utils import MODEL_WEIGHTS_S3_BUCKET, MODELS_ON_S3
|
||||
from vllm.transformers_utils.config import (
|
||||
get_model_path,
|
||||
is_interleaved,
|
||||
@@ -1126,15 +1125,6 @@ class EngineArgs:
|
||||
if check_gguf_file(self.model):
|
||||
self.quantization = self.load_format = "gguf"
|
||||
|
||||
# NOTE: This is to allow model loading from S3 in CI
|
||||
if (
|
||||
not isinstance(self, AsyncEngineArgs)
|
||||
and envs.VLLM_CI_USE_S3
|
||||
and self.model in MODELS_ON_S3
|
||||
and self.load_format == "auto"
|
||||
):
|
||||
self.model = f"{MODEL_WEIGHTS_S3_BUCKET}/{self.model}"
|
||||
|
||||
if self.disable_mm_preprocessor_cache:
|
||||
logger.warning(
|
||||
"`--disable-mm-preprocessor-cache` is deprecated "
|
||||
|
||||
Reference in New Issue
Block a user