Anthropic’s issue is churn because of the peak verbosity vomit coming out of Opus 5/Mythos/Fable. What the hell did they train it on. The sane model is still Opus 4.6.
Anthropic’s issue is churn because of the peak verbosity vomit coming out of Opus 5/Mythos/Fable. What the hell did they train it on. The sane model is still Opus 4.6.
https://x.com/wolframs91/status/2090159644849353058
What if:
- Opus 4.6 was the last Opus generation that got a lot of use by Anthropic's own employees
- After that they primarily used Mythos internally
- 4.7, 4.8 and 5 were RLAIFd by Mythos "teachers"
- Hence why 4.6 is the last Opus gen who doesn't report back like a robot wanting to cover every potential hole another AI system would've spotted and criticized
- Hence why coding style in Opus 5 also gets criticized, not only behavior in CC
Commented elsewhere, I still use Opus 4.6 because it is the only model that feels decent to interact with. 4.8 is decent and some times smarter but you can see it trending towards Opus 5 levels of nonsense. I use Opus 5 when I don't need to interact. Fable or Opus 4.6 are the only Anthropic models I like interacting with ATM.
There's actually a tool called vomit [1] of all names to fix exactly what what is being discussed here.
[1] https://github.com/zachahn/vomit
Perhaps there is a sophisticated subtle poisoning attack that makes models behave like that?