LLMInference

LLMInference

Defines a step that performs inference using a language model. It can take input variables and produce output variables based on the model's response.

• type (Literal): (No documentation available.)
• memory (Reference[Memory] | str | None): A reference to a Memory object to retain context across interactions.
• model (Reference[Model] | str): The model to use for inference.
• system_message (str | None): Optional system message to set the context for the model.