Users push for a full benchmark run on Opus 5 non-thinking as a budget pick

JasonBotterill · x · 2026-07-27

A user asks for a full benchmark suite on Opus 5 (non-thinking), arguing Anthropic skipped it for no good reason and that it looks promising as a budget option.

The post is essentially a call for a proper evaluation of the model’s non-reasoning mode, implying it could be useful if its benchmark numbers hold up.

Original post →

More from Models

Models channel →