What's your go-to LLM for coding?

Question

I've been using Gemini 3.1 Pro, mostly because I'd gotten used to Gemini in general, but it seems to be a relatively mediocre coder at best; it struggles, for example, on a ~600 LoC JS file, introducing approximately one new bug for every fixed one. That feels really unacceptable for a frontier model today, judging by all the praise I've been reading for Claude Code and Codex.If you've tried a few different frontier LLMs for coding recently, do you have any strong recommendations?

mwigdahl · Accepted Answer

Opus 4.7 in Claude Code. I've used GPT-5.5 in Codex and it's good also, but I have a pretty developed workflow in Claude Code so it's hard to tell if GPT-5.5 is actually rougher or if it just doesn't have the same tooling support I've built up in CC over the past 6 months.I should add that it's not perfect -- nothing in this space is -- but it to me has the best level of both consistency and quality of the setups I've tried.

Otternonsenz · Answer

And for the curiosity of those who are running open models locally, are there any open models that offer anything even looking like parity with commercial coding offerings? Or is there still more time for that to surface?