xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-06-06 19:29:09 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bugfix] Apply same sampling parameters for both n=1 and n>1 (#26005)

Browse Source 
Signed-off-by: Kenichi Maehashi <maehashi@preferred.jp>
Signed-off-by: yewentao256 <zhyanwentao@126.com>
This commit is contained in:
Kenichi Maehashi 2025-10-01 23:37:35 +09:00 committed by yewentao256
parent aeff0604bb
commit 0944358a90
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/v1/engine/async_llm.py
View File

@ -290,7 +290,7 @@ class AsyncLLM(EngineClient):
return queue return queue
# Fan out child requests (for n>1). # Fan out child requests (for n>1).
parent_request = ParentRequest(request_id, params) parent_request = ParentRequest(request_id, request.sampling_params)
for idx in range(params.n): for idx in range(params.n):
request_id, params = parent_request.get_child_info(idx) request_id, params = parent_request.get_child_info(idx)
child_request = request if idx == params.n - 1 else copy(request) child_request = request if idx == params.n - 1 else copy(request)