A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 23:30 UTC
[Bugfix] Handle content type with optional parameters (#13383)
Signed-off-by: Zifei Tong <zifeitong@gmail.com>
zifeitong committed
Feb 18, 2025 at 03:29 UTC
d3231cb43651c7d9f457e32708de3c22044c4446
1 file changed
+2
-1
vllm/entrypoints/openai/api_server.py
+2
-1
@@ -258,7 +258,8 @@ async def build_async_engine_client_from_engine_args(
258
259
async def validate_json_request(raw_request: Request):
260
content_type = raw_request.headers.get("content-type", "").lower()
261
- if content_type != "application/json":
261
+ media_type = content_type.split(";", maxsplit=1)[0]
262
+ if media_type != "application/json":
263
raise HTTPException(
264
status_code=HTTPStatus.UNSUPPORTED_MEDIA_TYPE,
265
detail="Unsupported Media Type: Only 'application/json' is allowed"