pub async fn process_batch_smart<T: ChatApi + ?Sized>(
api: &T,
requests: Vec<ChatCompletionRequest>,
concurrency_limit: Option<usize>,
) -> Result<Vec<Result<ChatCompletionResponse, AiLibError>>, AiLibError>
Expand description
智能批处理:根据请求类型和大小自动选择处理策略
Smart batch processing: automatically choose processing strategy based on request type and size