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 14, 05:30 UTC
[BugFix] enable DOTALL to match multi-line tool_call parameters in extract_tool_call_required_streaming (#24668)
Signed-off-by: Shijun Yin <shijun.yin@outlook.com>
Shijun Yin committed
Sep 17, 2025 at 17:21 UTC
2b856970313e80a649a573879d94e9d3430ba018
1 file changed
+1
-1
vllm/entrypoints/openai/serving_chat.py
+1
-1
@@ -418,7 +418,7 @@ class OpenAIServingChat(OpenAIServing):
418
if not function_name_returned:
419
# get partly generated arguments from the latest tool call
420
param_match = re.search(r'.*"parameters":\s*(.*)',
421
- current_text)
421
+ current_text, re.DOTALL)
422
arguments = param_match.group(1) if param_match else ""
423
arguments, _ = OpenAIServingChat._filter_delta_text(
424
arguments, previous_text)