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] gpt-oss container tool output bug (#25485)
Signed-off-by: Alec Solder <alecs@fb.com> Co-authored-by: Alec Solder <alecs@fb.com>
Alec S committed
Sep 23, 2025 at 16:43 UTC
c828d1bf987bbfe412c1fcd46736b1598d647053
1 file changed
+2
-1
vllm/entrypoints/harmony_utils.py
+2
-1
@@ -317,7 +317,8 @@ def parse_output_message(message: Message) -> list[ResponseOutputItem]:
317
)
318
output_items.append(response_item)
319
elif recipient is not None and (recipient.startswith("python")
320
- or recipient.startswith("browser")):
320
+ or recipient.startswith("browser")
321
+ or recipient.startswith("container")):
322
for content in message.content:
323
reasoning_item = ResponseReasoningItem(
324
id=f"rs_{random_uuid()}",