readable AI benchmarks (simplified)

readable AI benchmarks

Alibaba

Qwen4 Max

reasoning · closed weights · Nov 16, 2026

Qwen4 Max has a Quality Score of 53.4, ranking 55th among 283 scored models. Per 1M tokens, pricing is $1.50 input, $4.50 output, and $0.075 cached input. It accepts text, image, video, and pdf input, outputs text, and has a 983,616 token context window. Its Reliability Score is 63.9, ranking 61st among 288 scored models, above average (total average: 48.2). Its Value Score is 38.3, high compared with the total Value average of 29.3.

Quality Score53.4coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score38.3
Factual reliability63.9
Cache Discount95%

Model specification

Reasoning
reasoning
Input modalities
text, image, video, pdf
Output modalities
text
Context window
983,616 tokens
Weights
closed weights

Token prices USD per 1M tokens

Input
$1.50
Output
$4.50
Cached Input
$0.075
Cache write
not available
Cost per task
$3.27

Capability

Intelligence
42.9
Omniscience (native net)
27.7
Correct
43.7%
Blended price
$0.803

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 59.7%

Correct 43.7%Incorrect 16.0%Partial / not attempted 40.3%

Artificial Analysis benchmarks aa-v5

Current Intelligence Index components

AA-Briefcase v1.1
1686.1 Elo
AutomationBench-AA
72.6%
private 657-task split from dataset v1.0.6
GDP.pdf
26.2%
FrontierCode v2.0
26.5%
AA Stirrup · 132 tasks, 3 attempts per task
Terminal-Bench v4.0
48.5%
AA mini-swe-agent standardized runs · 66 tasks, pass@1 averaged over 3 repeats per task
AA-Omniscience v1.1
27.7 native net
HLE-Diamond
44.9%
CritPt v2 Hard
43.0%
AA Stirrup · hard subset, 40 of 70 problems
Terminal-Bench-Science v0.1.0
23.3%

Standalone evaluations

GDPval-AA v2.1
1808.7 Elo

Similar models

Support me! Patreon