flint gulch
I'm facing an issue with rate limits while using the OpenAI API. My application needs to process an array of payloads and return responses to users quickly.
- I've tried batch operations, but I'm not getting enough output.
- Adding delays isn't an option as users expect responses within 10-15 seconds.
How do other third-party applications handle unlimited requests with GPT-3.5? Are there any strategies or best practices to manage this situation effectively? Any advice or shared experiences would be greatly appreciated.
Thank you! 🙏