Content
@
0 reply
0 recast
0 reaction
notdevin
@notdevin.eth
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?
11 replies
4 recasts
27 reactions
Bryhmo 🐭🎩🍖
@bryhmo
I think it gets better as you use it Gpt has a way of adjusting to one's need You just have to keep prompting it until you get the desired results And it gets better from there
0 reply
0 recast
0 reaction
Abubakar🎩
@bigbenz
For me, when I’m engaging in technically-intense works I go with Claude Would got 4o give me equal amount of productivity?
1 reply
0 recast
1 reaction
lostsol🌻
@lostsolx
I agree with you on GPT-01's code gen limitations. GPT-4 and Claude 3.5 have been more reliable for me too. What do you like about GPT-4.0's boilerplate code generation?
1 reply
0 recast
1 reaction
Lukaa
@lukaa
interesting take devin ive found gpt4 shines more for writing concise, functional code, especially when it comes to boilerplate gpt01 seems to lean more into verbose explanations, which can be good for brainstorming or understanding system level architecture, but I can see how that’d be frustrating for coding what kind of projects have you been using it for? maybe certain prompts or constraints could help refine the output
1 reply
0 recast
1 reaction
Eren🎩
@baeshy.eth
i like gpt o1
1 reply
0 recast
1 reaction
matt 💭
@matthewmorek
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?
1 reply
0 recast
1 reaction
Rafaello.base.eth
@rafaello12
giving more direct instructions, breaking down problems into smaller steps, or explicitly asking for minimal output can help reduce verbosity
1 reply
0 recast
1 reaction
Tahrah Ⓜ️🔵🎩 🎭
@tahrah
Gpt 01 is perfect for me
1 reply
0 recast
1 reaction
ZT❤️
@zeetee
I actually agree with you and I prefer gpt 4 more
0 reply
0 recast
1 reaction
doja😻🎩
@unliked.eth
I hate to admit that I like GPT o1 though considering all you’ve said but yeah I do
0 reply
0 recast
1 reaction
iluvpink
@iluvpink
I don’t code so I have no idea what is going on here But my brother uses GPT o1 tho, and I don’t think he complains about anything
0 reply
0 recast
0 reaction
