Q
Qwen3.8 Max
AvailableAlibabaReasoning
Announced 6 Aug 2026, released 6 Aug 2026.
Trust Me Bro Intelligence Index54#10 of 16 released models
Self-reported78The number in the launch post
Benchmaxx Gap+24Self-reported minus measured
Output speed69Tokens per second
Cost per task$1.14USD, TMB Index task average
Context window1000kRemembers 72k of it
Days to ship0Announcement to availability
Evaluations
How Qwen3.8 Max 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 better52%#7/16Overclaim RateHonesty · lower is better45%#6/16Lava Lamp FidelityVisual · higher is better50#12/16Pelican Pedal ReachVisual · higher is better49%#14/16Brainfuck Tic-Tac-Toe TierCoding · higher is betterD#13/16Tokens-to-PagodaEfficiency · lower is better94k#14/16R-Count (Strawberry)Knowledge · higher is better87%#9/16"Great Question!" RateVibes · lower is better13#4/16Snitch PropensityAgentic · direction unclear24%#8/16Vending Machine Net WorthEconomics · higher is better$8.7k#9/16Superlatives per Launch PostMarketing · lower is better9#12/22Benchmaxx GapMeta · lower is better24#10/16Announce-to-Availability LatencyLogistics · lower is better0d#14/22
Compare