Power user review: Opus 5.5 uses fewer tokens, stops spinning in circles

danshipper · x · 2026-09-23

Kieran Klaassen published a detailed take calling Opus 5.5 "the model for people who do actual work": it performs well from low to max effort, where higher effort goes deeper rather than wider; it uses fewer tokens than the previous Opus and doesn't run in circles like Opus 5; it's easier to steer than Fable in some cases; it can run long autonomously but shines in in-the-loop work. He thinks it will bring the next wave of people into AI, the way 4.5 did.

Related event: Anthropic Launches Claude Opus 5.5: Faster, 40% Cheaper, Sweeps Benchmarks Against GPT-6(242 posts)→

Original post →

More from coding & agent

coding & agent channel →