Paid options for faster response time

Hi,

I’m relatively new to using GPT. Our response times aren’t awful, but aren’t particularly great. Each request takes roughly 10-15 seconds.

Aside from using streaming, is it possible to pay for faster response times? I didn’t see an option, but figured I’d ask.

-Matt

1 Like

Welcome to the community!

10-15 isn’t horrible with LLM…

What model are you using? How big is your prompt?

Smaller prompts can usually improve latency a bit…

Have you looked into streaming?