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 5, 22:15 UTC
[frontend] dump openai responses type by alias (#38262)
Signed-off-by: cjackal <44624812+cjackal@users.noreply.github.com>
cjackal committed
Mar 27, 2026 at 14:58 UTC
2babac0bed181c28b3b72aed97826a8ae96ae047
1 file changed
+2
-2
vllm/entrypoints/openai/responses/protocol.py
+2
-2
@@ -105,8 +105,8 @@ def serialize_message(msg):
105
elif hasattr(msg, "to_dict"):
106
return msg.to_dict()
107
else:
108
- # fallback to pyandic dump
109
- return msg.model_dump_json()
108
+ # fallback to pydantic dump
109
+ return msg.model_dump_json(by_alias=True)
110
111
112
def serialize_messages(msgs):