on_llm_end

Note:

This async callback also handles both run types. Async chat models start with on_chat_model_start, but there is no on_chat_model_end; completion is routed here for callback API compatibility.