Support

Settings

Theme

Explore themes

Mode

Library

Looks borrowed from benchmark sites and products. Themes with a light and a dark version follow your mode.

Bring your own theme

Upload a Priors theme file, or have your agent make one: copy the prompt, tell it the look you want, and upload what it returns. With the Priors skill installed it already knows the format.

Custom themes stay in this browser. They travel inside shared chart links and a page's theme, so others see them as you do.

For the builder, Explore and model pages on this device. Published pages keep the theme their owner picked.

Mode

Update notices

A note when a benchmark one of your indexes uses gets a new version, with a one-click upgrade.

Alibaba · released 2026-11-16

Qwen 4 Max

Measured on 57 benchmarks.

42.9

#16 of 73 models · best at xhigh

Strengths

  • Qwen 4 Max
  • Fable 5.2 (#1)
  • Field median
bestField medianFrontierCode 2.0-AATerminal-Benchv4-AAAutomationBench-AAAA-Briefcase v1.1GDP.pdfTerminal-Bench-Science-AACritPt v2 HardSubsetAA-Omnisciencev1.1 AccuracyHLE-DiamondAA-Omnisciencev1.1Non-hallucination
  1. FrontierCode 2.0-AA15%26.5%#21 of 73
  2. Terminal-Bench v4-AA10%48.5%#15 of 73
  3. AutomationBench-AA5%72.6%#6 of 73
  4. AA-Briefcase v1.115%1686#5 of 73
  5. GDP.pdf5%26.2%#11 of 73
  6. Terminal-Bench-Science-AA15%23.3%#16 of 73
  7. CritPt v2 Hard Subset10%43.0%#21 of 73
  8. AA-Omniscience v1.1 Accuracy10%43.7%#22 of 73
  9. HLE-Diamond10%44.9%#13 of 73
  10. AA-Omniscience v1.1 Non-hallucination5%71.6%#19 of 73

Bars show where Qwen 4 Max sits among the 73 ranked models on each benchmark (right is best); the tick is the field median.

By category

  • Qwen 4 Max
  • Fable 5.2 (#1)
  • Field median
p25p50p75topField medianCodingAgents & computeruseProfessional workScience & researchReasoning &puzzlesKnowledge & accuracyVisionLong documentsWriting & designHuman preference
  1. Coding7 benchmarks73%#10 of 73
  2. Agents & computer use20 benchmarks77%#7 of 73
  3. Professional work13 benchmarks80%#7 of 73
  4. Science & research3 benchmarks77%#9 of 72
  5. Reasoning & puzzles4 benchmarks69%#11 of 70
  6. Knowledge & accuracy2 benchmarks80%#8 of 69
  7. Vision2 benchmarks72%#11 of 56
  8. Long documents2 benchmarks91%#1 of 71
  9. Writing & design3 benchmarks58%#17 of 47
  10. Human preference2 benchmarks82%#5 of 47

Across 58 results on every current benchmark that measured Qwen 4 Max, not only this ranking’s. The percentage is its average standing among the 73 models in this ranking that took the same benchmarks: 100% beat them all, 0% trailed them all. A benchmark in two categories counts in both; a category needs two benchmarks.

Cost and speed

against 73 ranked models

Cost per task

#53 cheapest of 56

$3.3at xhigh

Output speed

#54 fastest of 61

53 tok/sat xhigh

Time to first token

#33 quickest of 61

5.3sat xhigh

Output tokens per task

#36 fewest of 61

105kat xhigh

API default effort: xhigh, per Artificial Analysis.

Token price

$0.133 per 1M tokens at API prices

97.0% cache read · 2.50% fresh input · 0.00% cache write · 0.50% output

Assumed mix: no measured T3 Code usage for this model.

Not measured for Qwen 4 Max: time per task.

This ranking’s benchmarks

10 of 10 measured

Cost per task is what a subscriber pays: the model’s plan price, a share of API price. Models no plan sells are at API price.

Agents & computer use

AutomationBench-AA

262 results
  1. #10xhigh of 26272.6%

Leader: Gemini 4 Carbon · 77.8%

Professional work

AA-Briefcase v1.1

267 results
  1. #8xhigh of 2671686

Leader: Sonnet 5.5 · 1816

Science & research

CritPt v2 Hard Subset

136 results
  1. #57xhigh of 13643.0%

Leader: Fable 5.2 · 76.5%

Coding

FrontierCode 2.0-AA

134 results
  1. #48xhigh of 13426.5%

Leader: Fable 5.2 · 49.1%

Professional work

GDP.pdf

260 results
  1. #39xhigh of 26026.2%

Leader: Gemini 4 Hydrogen · 32.8%

Knowledge & accuracy

HLE-Diamond

170 results
  1. #34xhigh of 17044.9%

Leader: Fable 5.2 · 80.2%

Knowledge & accuracy

AA-Omniscience v1.1 Accuracy

143 results
  1. #58xhigh of 14343.7%

Leader: Fable 5.2 · 79.1%

Knowledge & accuracy

AA-Omniscience v1.1 Non-hallucination

143 results
  1. #34xhigh of 14371.6%

Leader: Gemini 4 Carbon · 95.8%

Science & research

Terminal-Bench-Science-AA

145 results
  1. #44xhigh of 14523.3%

Leader: Fable 5.2 · 73.3%

Agents & computer use

Terminal-Bench v4-AA

258 results
  1. #34xhigh of 25848.5%

Leader: Kimi K3.1 · 67.2%

All 57 benchmarks

show 47 more