from_llm() — langchain Function Reference

Architecture documentation for the from_llm() function in base.py from the langchain codebase.

Function python AgentOrchestration ClassicChains

Entity Profile

AgentOrchestration→ ClassicChains→ from_llm() — langchain Function Reference

Dependency Diagram

graph TD
  37f412df_56f8_5f88_6665_6688ab4d38b7["from_llm()"]
  c827daf0_9d8e_4865_a678_af8144586d0e["FlareChain"]
  37f412df_56f8_5f88_6665_6688ab4d38b7 -->|defined in| c827daf0_9d8e_4865_a678_af8144586d0e
  style 37f412df_56f8_5f88_6665_6688ab4d38b7 fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

libs/langchain/langchain_classic/chains/flare/base.py lines 250–311

    def from_llm(
        cls,
        llm: BaseLanguageModel | None,
        max_generation_len: int = 32,
        **kwargs: Any,
    ) -> FlareChain:
        """Creates a FlareChain from a language model.

        Args:
            llm: Language model to use.
            max_generation_len: Maximum length of the generated response.
            kwargs: Additional arguments to pass to the constructor.

        Returns:
            FlareChain class with the given language model.
        """
        try:
            from langchain_openai import ChatOpenAI
        except ImportError as e:
            msg = (
                "OpenAI is required for FlareChain. "
                "Please install langchain-openai."
                "pip install langchain-openai"
            )
            raise ImportError(msg) from e
        # Preserve supplied llm instead of always creating a new ChatOpenAI.
        # Enforce ChatOpenAI requirement (token logprobs needed for FLARE).
        if llm is None:
            llm = ChatOpenAI(
                max_completion_tokens=max_generation_len,
                logprobs=True,
                temperature=0,
            )
        else:
            if not isinstance(llm, ChatOpenAI):
                msg = (
                    f"FlareChain.from_llm requires ChatOpenAI; got "
                    f"{type(llm).__name__}."
                )
                raise TypeError(msg)
            if not getattr(llm, "logprobs", False):  # attribute presence may vary
                msg = (
                    "Provided ChatOpenAI instance must be constructed with "
                    "logprobs=True for FlareChain."
                )
                raise ValueError(msg)
            current_max = getattr(llm, "max_completion_tokens", None)
            if current_max is not None and current_max != max_generation_len:
                logger.debug(
                    "FlareChain.from_llm: supplied llm max_completion_tokens=%s "
                    "differs from requested max_generation_len=%s; "
                    "leaving model unchanged.",
                    current_max,
                    max_generation_len,
                )
        response_chain = PROMPT | llm
        question_gen_chain = QUESTION_GENERATOR_PROMPT | llm | StrOutputParser()
        return cls(
            question_generator_chain=question_gen_chain,
            response_chain=response_chain,
            **kwargs,
        )