xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-10 23:55:19 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix: only response [DONE] once when streaming response. (#378)

Browse Source
This commit is contained in:
Ricardo Lu 2023-07-07 09:08:40 +08:00 committed by GitHub
parent 1c395b4eaa
commit b396cb4998
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
vllm/entrypoints/openai/api_server.py
View File

@ -269,7 +269,7 @@ async def create_chat_completion(raw_request: Request):
finish_reason=output.finish_reason, finish_reason=output.finish_reason,
) )
yield f"data: {response_json}\n\n" yield f"data: {response_json}\n\n"
yield "data: [DONE]\n\n" yield "data: [DONE]\n\n"
# Streaming response # Streaming response
if request.stream: if request.stream:
@ -465,7 +465,7 @@ async def create_completion(raw_request: Request):
finish_reason=output.finish_reason, finish_reason=output.finish_reason,
) )
yield f"data: {response_json}\n\n" yield f"data: {response_json}\n\n"
yield "data: [DONE]\n\n" yield "data: [DONE]\n\n"
# Streaming response # Streaming response
if stream: if stream: