logoalt Hacker News

colingauvinyesterday at 9:22 PM1 replyview on HN

Claude 4.5/4.6 - absolutely agree. Fable 5? From my (limited) testing, also reasonable to interact with.

Opus 4.7/4.8/5? Absolutely smug and antagonistic and preachy. I'm constantly fighting with it to stop fighting me and accept that I occasionally know better. It's really frustrating to spend so many tokens of such an expensive model arguing with it.


Replies

quadrupletoday at 1:02 PM

yeah, it's honestly amazing just how badly anthropic managed to screw up something in claude after 4.6. its night and day, and every time I see claude doing something stupid, i instantly realize I was accidentally on Opus 5