FakeMessagesListChatModel Class — langchain Architecture

Architecture documentation for the FakeMessagesListChatModel class in fake_chat_models.py from the langchain codebase.

Class python

Entity Profile

Dependency Diagram

graph TD
  82daee4e_6e0c_62d7_44e9_dea1c287f5ea["FakeMessagesListChatModel"]
  d009a608_c505_bd50_7200_0de8a69ba4b7["BaseChatModel"]
  82daee4e_6e0c_62d7_44e9_dea1c287f5ea -->|extends| d009a608_c505_bd50_7200_0de8a69ba4b7
  625e90ff_0acf_2872_ee23_0e50b0ab92ed["fake_chat_models.py"]
  82daee4e_6e0c_62d7_44e9_dea1c287f5ea -->|defined in| 625e90ff_0acf_2872_ee23_0e50b0ab92ed
  144796a8_e2cb_b1f5_10d6_fc85cfadd156["_generate()"]
  82daee4e_6e0c_62d7_44e9_dea1c287f5ea -->|method| 144796a8_e2cb_b1f5_10d6_fc85cfadd156
  6c353bb6_0e04_1e34_99ee_03670542e76e["_llm_type()"]
  82daee4e_6e0c_62d7_44e9_dea1c287f5ea -->|method| 6c353bb6_0e04_1e34_99ee_03670542e76e

Relationship Graph

Source Code

libs/core/langchain_core/language_models/fake_chat_models.py lines 21–52

class FakeMessagesListChatModel(BaseChatModel):
    """Fake chat model for testing purposes."""

    responses: list[BaseMessage]
    """List of responses to **cycle** through in order."""
    sleep: float | None = None
    """Sleep time in seconds between responses."""
    i: int = 0
    """Internally incremented after every model invocation."""

    @override
    def _generate(
        self,
        messages: list[BaseMessage],
        stop: list[str] | None = None,
        run_manager: CallbackManagerForLLMRun | None = None,
        **kwargs: Any,
    ) -> ChatResult:
        if self.sleep is not None:
            time.sleep(self.sleep)
        response = self.responses[self.i]
        if self.i < len(self.responses) - 1:
            self.i += 1
        else:
            self.i = 0
        generation = ChatGeneration(message=response)
        return ChatResult(generations=[generation])

    @property
    @override
    def _llm_type(self) -> str:
        return "fake-messages-list-chat-model"