Properly return error response in vertex_generate HTTP endpoint (#5956)

2025-05-01 11:48:58 -07:00
parent 6fc175968c
commit d33955d28a
2 changed files with 12 additions and 0 deletions
--- a/python/sglang/srt/entrypoints/http_server.py
+++ b/python/sglang/srt/entrypoints/http_server.py
@@ -675,6 +675,8 @@ async def vertex_generate(vertex_req: VertexGenerateReqInput, raw_request: Reque
        **(vertex_req.parameters or {}),
    )
    ret = await generate_request(req, raw_request)
+    if isinstance(ret, Response):
+        return ret
    return ORJSONResponse({"predictions": ret})