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

[Frontend] Use request id from header (#10968)

Browse Source 
Signed-off-by: Joe Runde <Joseph.Runde@ibm.com>
This commit is contained in:
Joe Runde
2024-12-09 22:46:29 -07:00
committed by GitHub
parent 391d7b2763
commit 980ad394a8
8 changed files with 27 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

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

@@ -305,7 +305,7 @@ async def health(raw_request: Request) -> Response:
async def tokenize(request: TokenizeRequest, raw_request: Request):
handler = tokenization(raw_request)
generator = await handler.create_tokenize(request)
generator = await handler.create_tokenize(request, raw_request)
if isinstance(generator, ErrorResponse):
return JSONResponse(content=generator.model_dump(),
status_code=generator.code)
@@ -319,7 +319,7 @@ async def tokenize(request: TokenizeRequest, raw_request: Request):
async def detokenize(request: DetokenizeRequest, raw_request: Request):
handler = tokenization(raw_request)
generator = await handler.create_detokenize(request)
generator = await handler.create_detokenize(request, raw_request)
if isinstance(generator, ErrorResponse):
return JSONResponse(content=generator.model_dump(),
status_code=generator.code)