Anthropic appears to be A/B testing reduced effort levels in Claude Code
A user on X reports that Anthropic appears to be A/B testing reduced effort levels in Claude Code. The user notes an update: the change is server-side, not in the app. Anthropic enrolls five sessions on Claude Code 2.1.236+ into an experiment that shrinks the effort scale. Older versions and Opus 5 are left alone. The user suggests it is probably an A/B test, so not everyone will see it. The user also notes that if "high" feels like "low" for you, that may be why. The source is a tweet, so details are limited, but the key points are that the experiment affects specific sessions and versions, and the effort scale is reduced.
// why it matters
This A/B test may change how Claude Code responds to effort settings for some users.