notdevin

GPT o1, so far, sucks for code as far as I can tell. It’s insanely verbose and goes off the rails just as fast, but now for longer, than previous models. It has been awesome for conceptual brainstorming and system architecture. I much prefer Claude 3.5 or gpt 4, I’ll take 4o for boiler plate specifically

Anyone else writing code having better experiences with o1 preview?

@backpack put a bank in your pocket and use your money whenever, wherever, however without permission

/backpack
/notdevin
icebreaker.xyz/notdevin

GPT-4o still works better for code IMO, need to keep it in check with custom guidance but otherwise works well.

Claude 3.5 Sonnet hallucinates a little and goes off rails but it’s meant to, so I use it for generating good commit messages.

Any other models you’ve tried and can recommend to try?

✳︎ design engineer → @ponder 
✳︎ explorer of humanism • petrolhead/gearhead
✳︎ React/SwiftUI
✳︎ bento.me/matthew

I’ve only had time to test out the common models for code yet, I’d like to get to a llama model but not while I’m on tough deadlines