search

BerriAI / reliableGPT

Get 100% uptime, reliability from OpenAI. Handle Rate Limit, Timeout, API, Keys Errors
MIT License
623 stars 33 forks source link

Should I add parallel request functionality #8

Open George3d6 opened 1 year ago

George3d6 commented 1 year ago

Is this something you'd want PRed into the library? Ideally with both an asyncio and a non-asyncio interface.

Can't 100% promise I'll do it since doing it right will take some time, but seems like almost everyone using this library will have their own parallelization wrapper on top, might as well add it to it

krrishdholakia commented 1 year ago

That's a brilliant idea @George3d6. We're in the process of making this push. Will @ you in the PR for review.

Want to talk through the implementation?

Attaching my calendly if that helps - https://calendly.com/kdholakia/30min

krrishdholakia commented 1 year ago

https://github.com/BerriAI/reliableGPT/commit/a41b121f6a00a3bf2b88ff814cce818efcd11288

https://github.com/BerriAI/reliableGPT/pull/9

^ working on a colab notebook to show implementation @George3d6