Closed rodonguyen closed 7 months ago
I will write a test to test the ability of litellm in handling multiple request
I don't think we put a high priority on using OpenAI to response to the users. Remember OpenAI resource is not something we can completely use for free and is at the good will of the donated key.
It is fine to keep the response delayed and use the local LLM model as I can try getting us more funding on that aspect to improve the model responsiveness. I do not have that luxury with OpenAI models.
await ctx.defer()
) hay nhiều requests làm đơ bot