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

[Main][Refactor]Change ASCEND_QUATIZATION_METHOD to ASCEND_QUANTIZATION_METHOD (#2517)

Browse Source 
### What this PR does / why we need it?
The constant ASCEND_QUATIZATION_METHOD in vllm_ascend/utils.py is
misspelled and should be corrected to ASCEND_QUANTIZATION_METHOD.

### Does this PR introduce _any_ user-facing change?
No

### How was this patch tested?
CI passed with new added/existing test.

- vLLM version: v0.10.1.1
- vLLM main:
c9abb10489

Signed-off-by: zhanghaiwen <zhanghaiwen@cmss.chinamobile.com>
Co-authored-by: zhanghaiwen <zhanghaiwen@cmss.chinamobile.com>
This commit is contained in:
zhanghw0354
2025-08-26 09:06:16 +08:00
committed by GitHub
parent 21b5727f9a
commit b3fdd78a6b
5 changed files with 16 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
vllm_ascend/platform.py
View File

@@ -28,7 +28,7 @@ from vllm.platforms import Platform, PlatformEnum
from vllm_ascend.ascend_config import (check_ascend_config, get_ascend_config,
init_ascend_config)
from vllm_ascend.utils import (ASCEND_QUATIZATION_METHOD, is_310p,
from vllm_ascend.utils import (ASCEND_QUANTIZATION_METHOD, is_310p,
update_aclgraph_sizes)
if TYPE_CHECKING:
@@ -50,7 +50,7 @@ class NPUPlatform(Platform):
device_control_env_var: str = "ASCEND_RT_VISIBLE_DEVICES"
dispatch_key: str = "PrivateUse1"
supported_quantization: list[str] = [ASCEND_QUATIZATION_METHOD]
supported_quantization: list[str] = [ASCEND_QUANTIZATION_METHOD]
def is_sleep_mode_available(self) -> bool:
return True
@@ -70,8 +70,8 @@ class NPUPlatform(Platform):
quant_action = parser._option_string_actions.get('--quantization')
if quant_action and hasattr(quant_action,
'choices') and quant_action.choices:
if ASCEND_QUATIZATION_METHOD not in quant_action.choices:
quant_action.choices.append(ASCEND_QUATIZATION_METHOD)
if ASCEND_QUANTIZATION_METHOD not in quant_action.choices:
quant_action.choices.append(ASCEND_QUANTIZATION_METHOD)
from vllm_ascend.quantization.quant_config import \
AscendQuantConfig # noqa: F401