Lately, I've been using 4o for my coding questions, particularly regarding best practices and problem-solving. However, I've noticed that the answers seem pretty limited and often overlook various perspectives on the issue. I'm wondering if this is a general issue with 4o or something that's just happened recently. Should I switch to using o3 instead for coding advice? What has your experience been like with these models?
1 Answer
If you can access the API, I've found that GPT 4.1 performs significantly better. For coding questions that aren't too basic, I'd recommend sticking with the o4-mini-high or o3.
What's the 'personality' of 4.1 like? Does it come off as more objective like the reasoning models or more personal?