Skip to content

vllm.entrypoints.serve.utils.tool_calls_utils

maybe_filter_parallel_tool_calls 

maybe_filter_parallel_tool_calls(
    choice: _ChatCompletionResponseChoiceT,
    request: ChatCompletionRequest,
) -> _ChatCompletionResponseChoiceT

Filter to first tool call only when parallel_tool_calls is False.

Source code in vllm/entrypoints/serve/utils/tool_calls_utils.py 
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
def maybe_filter_parallel_tool_calls(
    choice: _ChatCompletionResponseChoiceT, request: ChatCompletionRequest
) -> _ChatCompletionResponseChoiceT:
    """Filter to first tool call only when parallel_tool_calls is False."""

    if request.parallel_tool_calls:
        return choice

    if isinstance(choice, ChatCompletionResponseChoice) and choice.message.tool_calls:
        choice.message.tool_calls = choice.message.tool_calls[:1]
    elif (
        isinstance(choice, ChatCompletionResponseStreamChoice)
        and choice.delta.tool_calls
    ):
        choice.delta.tool_calls = [
            tool_call for tool_call in choice.delta.tool_calls if tool_call.index == 0
        ]

    return choice