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 7, 05:15 UTC
[CI] Fix pre-commit mypy issue in main (#36049)
Wentao Ye committed
Mar 4, 2026 at 21:13 UTC
2f4226fe5280b60c47b4f6f01d9b18ac9cda2038
1 file changed
+3
-2
vllm/tool_parsers/hermes_tool_parser.py
+3
-2
@@ -329,11 +329,12 @@ class Hermes2ProToolParser(ToolParser):
329
logger.debug("unable to parse JSON")
330
return None
331
332
+ if current_tool_call is None:
333
+ return None
334
+
335
# case - we haven't sent the tool name yet. If it's available, send
336
# it. otherwise, wait until it's available.
337
if not self.current_tool_name_sent:
335
- if current_tool_call is None:
336
- return None
338
function_name: str | None = current_tool_call.get("name")
339
if function_name:
340
self.current_tool_name_sent = True