Qwen4 Max uses 39% fewer output tokens while scoring 9 points higher
ArtificialAnlys · x · 2026-11-16
Artificial Analysis charts show Qwen4 Max averages 105k output tokens per Intelligence Index task (56k reasoning), 39% fewer than Qwen3.8 Max's 171k, while scoring 9 points higher on the index. For comparison, Kimi K3.1 Pro uses 72k and Tencent Hy4 78k. With most models' output dominated by reasoning tokens, Qwen4 Max improving scores while thinking less marks a notable gain in reasoning efficiency.
More from Models
- Qwen4 Max Scores 43 on Intelligence Index With 984k Context, Artificial Analysis Finds — ArtificialAnlys · 2026-11-16
- Qwen4 Max per-eval breakdown: strong on knowledge work, lagging in coding — ArtificialAnlys · 2026-11-16
- How Qwen4 Max cut cost per task 67%: cheaper cache hits and 39% fewer tokens — ArtificialAnlys · 2026-11-16
- Qwen4 Max scores 43 on AA Intelligence Index, 67% cheaper per task than predecessor — ArtificialAnlys · 2026-11-16
- Qwen4-Max debuts at #14 on Agent Arena, making Alibaba the #5 lab overall — Alibaba_Qwen · 2026-11-16
- New MoE base claims win over Qwen3.8-Max with 50% fewer tokens, met with skepticism — teortaxesTex · 2026-11-16