(input_file, output_file, error_file, limit=1000, num_threads=4)

(model, tokenizer, prompt, max_length=8192)

()

(example)

( message: str, chat_history: list[tuple[str, str]], system_prompt: str, max_n