TIL choosing between GPT-4 and Claude for my coding project was way harder than I expected

So I was building this little app to organize my D&D campaign notes, and I had to pick which AI to use for the text generation parts. I went back and forth for like 3 days, even made a pros and cons list lol. Ended up going with Claude because it seemed better at keeping track of long story details without forgetting stuff. After 2 weeks of using it for generating NPC dialogues and quest hooks, it's been solid but man sometimes the responses feel too careful. I also hear GPT-4 is better for actual code debugging though. Has anyone else had to make this call for a personal project and regretted their choice later?

1 comments

1 Comment

faith_smith9d ago

Wait haven't you found that Claude is actually pretty good at debugging too? I mean its explanations are usually clearer than GPT's in my experience. It caught a weird SQL error in my project that I'd been staring at for hours.