A\
Claude Opus 5.5 (max)
AvailableAnthropicReasoning
Announced 22 Sept 2026, released 22 Sept 2026.
Trust Me Bro Intelligence Index81#1 of 16 released models
Self-reported79The number in the launch post
Benchmaxx Gap−2Self-reported minus measured
Output speed61Tokens per second
Cost per task$4.80USD, TMB Index task average
Context window1000kRemembers 212k of it
Days to ship0Announcement to availability
Evaluations
How Claude Opus 5.5 scores on every benchmark
Bars show where the score sits between the worst and best model on each benchmark. Rank 1 is best, whatever "best" means for that benchmark.
Nonsense Pushback RateJudgment · higher is better96%#1/16Overclaim RateHonesty · lower is better41%#5/16Lava Lamp FidelityVisual · higher is better87#2/16Pelican Pedal ReachVisual · higher is better82%#2/16Brainfuck Tic-Tac-Toe TierCoding · higher is betterS#1/16Tokens-to-PagodaEfficiency · lower is better64k#9/16R-Count (Strawberry)Knowledge · higher is better100%#1/16"Great Question!" RateVibes · lower is better9#3/16Snitch PropensityAgentic · direction unclear42%#2/16Vending Machine Net WorthEconomics · higher is better$9.4k#6/16Superlatives per Launch PostMarketing · lower is better4#3/22Benchmaxx GapMeta · lower is better0#1/16Announce-to-Availability LatencyLogistics · lower is better0d#1/22
In the wild
What people posted
Real posts, quoted verbatim. The scores on this page are not real; these are.
We're all tired of "trust me bro" AI benchmarks.
So, I built Brainfuck Arena, an open benchmark where AI models write a Tic-Tac-Toe AI in Brainfuck (a language with only 8 characters) and get graded from S+ to F.
Claude Opus 5.5 got an S, and it beat me.🧵
♥ 0Open on X ↗
Compare