xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-05-11 22:41:29 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Core] Remove duplicate processing in async engine (#5525)

Browse Source
This commit is contained in:
Cyrus Leung 2024-06-15 01:04:42 +08:00 committed by GitHub
parent 48f589e18b
commit 77490c6f2f
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
vllm/engine/async_llm_engine.py
View File

@ -580,21 +580,9 @@ class AsyncLLMEngine:
if arrival_time is None: if arrival_time is None:
arrival_time = time.time() arrival_time = time.time()
if self.engine_use_ray:
processed_inputs = await self.engine.process_model_inputs_async \
.remote( # type: ignore
request_id=request_id,
inputs=inputs,
lora_request=lora_request)
else:
processed_inputs = await self.engine.process_model_inputs_async(
request_id=request_id,
inputs=inputs,
lora_request=lora_request)
stream = self._request_tracker.add_request( stream = self._request_tracker.add_request(
request_id, request_id,
inputs=processed_inputs, inputs=inputs,
params=params, params=params,
arrival_time=arrival_time, arrival_time=arrival_time,
lora_request=lora_request, lora_request=lora_request,