Function chat_completion_handler

llama_cpp/llama_chat_format.py:613–755

(
        *,
        llama: llama.Llama,
        messages: List[llama_types.ChatCompletionRequestMessage],
        functions: Optional[List[llama_types.ChatCompletionFunction]] = None,
        function_call: Optional[llama_types.ChatCompletionRequestFunctionCall] = None,
        tools: Optional[List[llama_types.ChatCompletionTool]] = None,
        tool_choice: Optional[llama_types.ChatCompletionToolChoiceOption] = None,
        temperature: float = 0.2,
        top_p: float = 0.95,
        top_k: int = 40,
        min_p: float = 0.05,
        typical_p: float = 1.0,
        stream: bool = False,
        stop: Optional[Union[str, List[str]]] = [],
        seed: Optional[int] = None,
        response_format: Optional[
            llama_types.ChatCompletionRequestResponseFormat
        ] = None,
        max_tokens: Optional[int] = None,
        presence_penalty: float = 0.0,
        frequency_penalty: float = 0.0,
        repeat_penalty: float = 1.1,
        tfs_z: float = 1.0,
        mirostat_mode: int = 0,
        mirostat_tau: float = 5.0,
        mirostat_eta: float = 0.1,
        model: Optional[str] = None,
        logits_processor: Optional[llama.LogitsProcessorList] = None,
        grammar: Optional[llama.LlamaGrammar] = None,
        logit_bias: Optional[Dict[str, float]] = None,
        logprobs: Optional[bool] = None,
        top_logprobs: Optional[int] = None,
        **kwargs,  # type: ignore
    )

    chat_formatter: ChatFormatter,
) -> LlamaChatCompletionHandler:
    def chat_completion_handler(
        *,
        llama: llama.Llama,
        messages: List[llama_types.ChatCompletionRequestMessage],
        functions: Optional[List[llama_types.ChatCompletionFunction]] = None,
        function_call: Optional[llama_types.ChatCompletionRequestFunctionCall] = None,
        tools: Optional[List[llama_types.ChatCompletionTool]] = None,
        tool_choice: Optional[llama_types.ChatCompletionToolChoiceOption] = None,
        temperature: float = 0.2,
        top_p: float = 0.95,
        top_k: int = 40,
        min_p: float = 0.05,
        typical_p: float = 1.0,
        stream: bool = False,
        stop: Optional[Union[str, List[str]]] = [],
        seed: Optional[int] = None,
        response_format: Optional[
            llama_types.ChatCompletionRequestResponseFormat
        ] = None,
        max_tokens: Optional[int] = None,
        presence_penalty: float = 0.0,
        frequency_penalty: float = 0.0,
        repeat_penalty: float = 1.1,
        tfs_z: float = 1.0,
        mirostat_mode: int = 0,
        mirostat_tau: float = 5.0,
        mirostat_eta: float = 0.1,
        model: Optional[str] = None,
        logits_processor: Optional[llama.LogitsProcessorList] = None,
        grammar: Optional[llama.LlamaGrammar] = None,
        logit_bias: Optional[Dict[str, float]] = None,
        logprobs: Optional[bool] = None,
        top_logprobs: Optional[int] = None,
        **kwargs,  # type: ignore
    ) -> Union[
        llama_types.CreateChatCompletionResponse,
        Iterator[llama_types.CreateChatCompletionStreamResponse],
    ]:
        result = chat_formatter(
            messages=messages,
            functions=functions,
            function_call=function_call,
            tools=tools,
            tool_choice=tool_choice,
            **kwargs,
        )
        prompt = llama.tokenize(
            result.prompt.encode("utf-8"),
            add_bos=not result.added_special,
            special=True,
        )
        if result.stop is not None:
            stop = [] if stop is None else [stop] if isinstance(stop, str) else stop
            rstop = result.stop if isinstance(result.stop, list) else [result.stop]
            stop = stop + rstop

        stopping_criteria = None
        if result.stopping_criteria is not None: