Anthropic appears to be A/B testing reduced effort levels in Claude Code
Users have discovered that Anthropic appears to be A/B testing a change in its Claude Code tool that reduces the 'effort' level of the AI model. This has caused frustration among developers who feel the model's performance has degraded without clear communication from the company.
Why it matters
This highlights the lack of transparency in AI product updates and the reliance developers have on consistent model behavior for professional workflows.
🥔🥔🥔 on X: "update: it's server-side, not the app
anthropic enrols fable 5 sessions on claude code 2.1.236+ into an experiment that shrinks the effort scale, older versions and opus 5 are left alone
probably an a/b test, so not everyone will see it
if "high" feels like "low" for you, you're in the test group
holy fuck anthropic, you guys are unbearable sometimes
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in