EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[CI] upgrade to vllm 0.9.0 (#959)

Browse Source 
Upgrade to vllm 0.9.0.
0.8.5 will not be supported any more.

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
This commit is contained in:
wangxiyuan
2025-05-28 21:18:41 +08:00
committed by GitHub
parent e2a0c19cea
commit f6e5decc10
16 changed files with 79 additions and 146 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
vllm_ascend/worker/model_runner.py
View File

@@ -64,8 +64,6 @@ from vllm.worker.model_runner_base import (
_init_attn_metadata_from_tensor_dict,
_init_sampling_metadata_from_tensor_dict)
from vllm_ascend.utils import vllm_version_is
if TYPE_CHECKING:
from vllm.attention.backends.abstract import AttentionBackend
@@ -1017,10 +1015,8 @@ class NPUModelRunnerBase(ModelRunnerBase[TModelInputForNPU]):
pattern: Optional[str] = None,
max_size: Optional[int] = None,
) -> None:
if vllm_version_is("0.8.5") or vllm_version_is("0.8.5.post1"):
from vllm.model_executor.model_loader.loader import ShardedStateLoader # type: ignore[import] # isort: skip # noqa
else:
from vllm.model_executor.model_loader import ShardedStateLoader
from vllm.model_executor.model_loader import ShardedStateLoader
ShardedStateLoader.save_model(
self.model,
path,
@@ -1032,12 +1028,9 @@ class NPUModelRunnerBase(ModelRunnerBase[TModelInputForNPU]):
self,
tensorizer_config: TensorizerConfig,
) -> None:
if vllm_version_is("0.8.5") or vllm_version_is("0.8.5.post1"):
from vllm.model_executor.model_loader.loader import \
TensorizerLoader # type: ignore # noqa
else:
from vllm.model_executor.model_loader import \
TensorizerLoader # type: ignore # noqa
from vllm.model_executor.model_loader import \
TensorizerLoader # type: ignore # noqa
TensorizerLoader.save_model(
self.model,
tensorizer_config=tensorizer_config,