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 4, 23:45 UTC
fix: Do not make function calls when request has no tools for /v1/responses (#40314)
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
Yuan Tang committed
Apr 20, 2026 at 00:17 UTC
67ed01c353f522f3265b95f8a01295f1d99d2969
1 file changed
+1
-1
vllm/entrypoints/openai/responses/utils.py
+1
-1
@@ -264,7 +264,7 @@ def convert_tool_responses_to_completions_format(tool: dict) -> dict:
264
def construct_tool_dicts(
265
tools: list[Tool], tool_choice: ToolChoice
266
) -> list[dict[str, Any]] | None:
267
- if tools is None or (tool_choice == "none"):
267
+ if not tools or (tool_choice == "none"):
268
tool_dicts = None
269
else:
270
tool_dicts = [