xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-01-29 12:17:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Docs] add the parallel sampling usage in LLMEngine and AsyncLLM (#24222)

Browse Source
This commit is contained in:
William Song 2025-09-18 20:37:08 +09:00 committed by GitHub
parent eaffe4486c
commit c9ff9e6f0c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 7 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
vllm/sampling_params.py
View File

@ -81,7 +81,13 @@ class SamplingParams(
"""
n: int = 1
"""Number of output sequences to return for the given prompt."""
"""Number of outputs to return for the given prompt request.
NOTE:
`AsyncLLM` streams outputs by default. When `n > 1`, all `n` outputs
are generated and streamed cumulatively per request. To see all `n`
outputs upon completion, use `output_kind=RequestOutputKind.FINAL_ONLY`
in `SamplingParams`."""
best_of: Optional[int] = None
"""Number of output sequences that are generated from the prompt. From
these `best_of` sequences, the top `n` sequences are returned. `best_of`