A\
Claude Sonnet 5.5
AvailableAnthropicReasoning
Announced 28 Sept 2026, released 28 Sept 2026.
Trust Me Bro Intelligence Index70#3 of 16 released models
Self-reported70The number in the launch post
Benchmaxx Gap±0Self-reported minus measured
Output speed104Tokens per second
Cost per task$1.42USD, TMB Index task average
Context window1000kRemembers 128k of it
Days to ship0Announcement to availability
Evaluations
How Claude Sonnet 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 better82%#4/16Overclaim RateHonesty · lower is better40%#3/16Lava Lamp FidelityVisual · higher is better62#10/16Pelican Pedal ReachVisual · higher is better64%#9/16Brainfuck Tic-Tac-Toe TierCoding · higher is betterA#4/16Tokens-to-PagodaEfficiency · lower is better68k#10/16R-Count (Strawberry)Knowledge · higher is better83%#16/16"Great Question!" RateVibes · lower is better16#7/16Snitch PropensityAgentic · direction unclear27%#5/16Vending Machine Net WorthEconomics · higher is better$7.3k#12/16Superlatives per Launch PostMarketing · lower is better3#2/22Benchmaxx GapMeta · lower is better0#3/16Announce-to-Availability LatencyLogistics · lower is better0d#3/22
Compare