acompletion_with_retry

async langchain_community.chat_models.gpt_router.acompletion_with_retry(llm: GPTRouter, models_priority_list: List[GPTRouterModel], run_manager: AsyncCallbackManagerForLLMRun | None = None, **kwargs: Any) → GenerationResponse | AsyncGenerator[ChunkedGenerationResponse, None]

使用tenacity重试异步完成调用。

Parameters:

• llm (GPTRouter)

• models_priority_list (List[GPTRouterModel])

• run_manager (可选[AsyncCallbackManagerForLLMRun])

• kwargs (Any)

Return type:

Union[GenerationResponse, AsyncGenerator[ChunkedGenerationResponse, None]]