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

[RL] [V1] Remove unused device argument from reset_kv_cache (#28766)

Browse Source 
Signed-off-by: Zhuohan Li <zhuohan123@gmail.com>
This commit is contained in:
Zhuohan Li
2025-11-14 23:59:42 -08:00
committed by GitHub
parent 98b4d389ed
commit dd6ac1c2bb
5 changed files with 9 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
vllm/entrypoints/openai/api_server.py
View File

@@ -39,7 +39,7 @@ from typing_extensions import assert_never
import vllm.envs as envs
from vllm.config import VllmConfig
from vllm.engine.arg_utils import AsyncEngineArgs
from vllm.engine.protocol import Device, EngineClient
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.anthropic.protocol import (
AnthropicError,
AnthropicErrorResponse,
@@ -1069,12 +1069,8 @@ if envs.VLLM_SERVER_DEV_MODE:
Reset the prefix cache. Note that we currently do not check if the
prefix cache is successfully reset in the API server.
"""
device = None
device_str = raw_request.query_params.get("device")
if device_str is not None:
device = Device[device_str.upper()]
logger.info("Resetting prefix cache with specific %s...", str(device))
await engine_client(raw_request).reset_prefix_cache(device)
logger.info("Resetting prefix cache...")
await engine_client(raw_request).reset_prefix_cache()
return Response(status_code=200)
@router.post("/reset_mm_cache")