I want to set up API request limit, what should I do?

[dangyuuki123]

Hi, my Ai system can't call too much, I tried to do sleep for asyncio but it doesn't work, how can I limit the calling speed like microsoft's graphrag?

[rangehow]

see *_max_async

nano-graphrag/nano_graphrag/graphrag.py

Lines 112 to 115 in 7a5686c

	best_model_max_async: int = 16
	cheap_model_func: callable = gpt_4o_mini_complete
	cheap_model_max_token_size: int = 32768
	cheap_model_max_async: int = 16