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 13, 13:15 UTC
[Frontend][2/n] remove empty content from _parse_tool_calls_from_content (#28331)
Signed-off-by: Andrew Xia <axia@fb.com> Co-authored-by: Andrew Xia <axia@fb.com>
Andrew Xia committed
Nov 10, 2025 at 14:07 UTC
4b94ed8f928533b1f7c3a0293790ccb592b49f1a
1 file changed
+2
vllm/entrypoints/openai/serving_engine.py
+2
@@ -1375,6 +1375,8 @@ class OpenAIServing:
1375
for tool_call in tool_call_info.tool_calls
1376
)
1377
content = tool_call_info.content
1378
+ if content and content.strip() == "":
1379
+ content = None
1380
else:
1381
# No tool calls.
1382
return None, content