Anthropic Engineer Addresses Claude Opus 5 Complaints
Users report Claude Opus 5 feels inconsistent and lazy compared to version 4.6.
TLDR
Kimmonismus posted that Claude Code appeared to run high reasoning effort at a lower level than before, making Fable feel dumber. Anthropic engineer Thariq replied that nothing changed on the models and the display was a numerical mapping issue with no performance regression. Kimmonismus added that Opus 5 repeatedly feels like a downgrade, producing sloppy mistakes and ignoring instructions. Thariq acknowledged the model is spiky and stated the team is working to make outputs consistent and warm while rolling out easier feedback tools.
Combined views
1.1M
10 Sources, first seen ago
Sources
CH 
Chubby♨️@kimmonismusWtf Anthropic: If Fable has felt noticeably dumber this week, it’s because Claude Code has apparently been running “high” reasoning effort at just 10/100, the same level that was „low“ before. They dumbed down fable without telling it seems.…
- likes: 2.4K
- replies: 185
- bookmarks: 395
- reposts: 101
CH Chubby♨️@kimmonismusAnthropic Thariq says Claude Code’s “high = 10” output was a numerical mapping issue, not a stealth downgrade. “High” is apparently still high, and internal evals show no performance regression. Thanks for the clarification. But I still have to say that most models currently…
- likes: 622
- replies: 46
- bookmarks: 73
- reposts: 21
CH Chubby♨️@kimmonismusWith all due respect to Boris Cherny, this response is part of the problem. The complaints about Opus 5 have become pretty consistent: it’s lazy, sloppy, and verbose. That matches my own experience. Responding that people simply haven’t understood the right use case misses the…
- likes: 2K
- replies: 152
- bookmarks: 197
- reposts: 56