4 ms·
Me too. Claude Opus 5's English is insufferable. Opus 4.6-4.8 was more reasonable. I've moved to Codex 5.6-Sol. Much saner English, much better at execution, a
by wenc 4d ago
Me too. Claude Opus 5's English is insufferable. Opus 4.6-4.8 was more reasonable.
I've moved to Codex 5.6-Sol. Much saner English, much better at execution, and gets stuff done in a matter-of-factly kind of way (Claude Code is a mess these days -- it gets things wrong and goes around in circles).
But I'm harness agnostic and am not locked in. I just keep my issues in Kata Tracker (https://www.katatracker.com/ https://www.katatracker.com/) and switch harnesses/model when I need to.
Being loyal to a particular model/harness seem unwise to me.
- amelius 4d ago> Me too. Claude Opus 5's English is insufferable. Opus 4.6-4.8 was more reasonable. I suspect this degradation is happening because the AI labs are using the LLM's output to feedback into the input, to create a thinking loop, and they're optimizing that.