Benchmarks · economics
GDPval-AA v2.1
GDPval-AA: economically valuable knowledge-work tasks across occupations, graded pairwise by AA; reported as Elo.
122 models · 217 results · data as of 2026-11-06 · artificialanalysis.ai ↗
Leaderboard
Opus 5.5 max
More from this benchmark
5 published measuresScore vs any published measure5 measures: sub-scores, splits, counts
86 more models under All models in settings.
Breakdown
darker is better within each column · click a column to sort| Model | ||||
|---|---|---|---|---|
| Opus 5.5 max | 3,793,360,879 | 14,463,522 | 26,714,602 | 3,726,668,094 |
| Kimi K3.1 max | 1,675,339,789 | 9,256,133 | 9,147,191 | 1,636,749,817 |
| Sonnet 5.5 max | 5,359,584,001 | 17,678,286 | 45,125,897 | 5,267,664,250 |
| Muse Wonder (Max) | 1,704,594,886 | 12,871,718 | 3,784,712 | 1,668,271,740 |
| Fable 5.1 max | 1,596,487,185 | 12,011,491 | 11,206,347 | 1,556,810,142 |
| DeepSeek V4.1 Flash 1017 max | 2,319,510,948 | 13,369,820 | 17,476,407 | 2,274,555,577 |
| Gemini 4 Carbon high | 570,714,529 | 9,820,463 | 7,629,530 | 539,377,964 |
| Opus 5 max | 1,443,328,948 | 12,091,348 | 9,300,465 | 1,407,667,015 |
| Grok 4.7 xhigh | 1,485,725,401 | 10,183,729 | 19,659,252 | 1,439,704,229 |
| MiMo V2.6 Pro | 911,856,109 | 9,528,936 | 9,817,589 | 882,066,968 |
| Muse Spark 1.3 max | 1,627,438,075 | 12,246,300 | 6,542,989 | 1,593,843,633 |
| GLM 5.3 max | 1,388,705,752 | 10,535,856 | 10,237,142 | 1,356,419,221 |
| Grok 4.6 xhigh | 695,339,345 | 8,679,385 | 4,836,099 | 668,009,169 |
| GLM 5.3 Flash max | 1,803,639,504 | 11,811,033 | 15,883,478 | 1,762,577,854 |
| Qwen 3.8 Flash | 2,430,846,723 | 14,616,659 | 19,219,025 | 2,375,191,777 |
| Haiku 5.5 max | 1,561,999,035 | 10,545,758 | 38,368,068 | 1,499,434,173 |
| Gemini 4 Argon high | 612,310,113 | 10,489,845 | 5,915,773 | 583,164,626 |
| Inkling 2 Small | 1,588,423,870 | 11,921,354 | 9,086,898 | 1,556,316,344 |
| Ling 3.1 Flash | 2,936,458,201 | 16,669,027 | 14,833,747 | 2,887,461,658 |
| Fable 5 max, fallback | 589,472,699 | 6,892,831 | 9,405,110 | 563,987,746 |
| GPT-5.6 Sol max | 787,741,056 | 4,988,962 | 3,939,218 | 765,724,742 |
| Qwen 3.8 2.4T-A95B | 1,278,900,227 | 9,641,779 | 9,408,962 | 1,249,681,811 |
| MiMo V2.6 Flash | 1,393,234,667 | 7,961,151 | 14,349,769 | 1,358,798,520 |
| Qwen 3.8 Max | 1,271,929,770 | 9,399,900 | 8,604,480 | 1,243,085,741 |
| GPT-6.1 Sol max | 374,392,193 | 4,267,790 | 5,818,492 | 352,441,989 |
All results
217 rows| # | Model | Effort | GDPval-AA v2.1 | Harness |
|---|---|---|---|---|
| 1 | Opus 5.5 | max | 1866 ±26 | Stirrup |
| 2 | Kimi K3.1 | max | 1858 ±28 | Stirrup |
| 3 | Sonnet 5.5 | max | 1838 ±23 | Stirrup |
| 4 | Opus 5.5 | xhigh | 1835 ±23 | Stirrup |
| 5 | Muse Wonder (Max) | max | 1770 ±23 | Stirrup |
| 6 | Fable 5.1 | max | 1763 ±19 | Stirrup |
| 7 | DeepSeek V4.1 Flash 1017 | max | 1746 ±21 | Stirrup |
| 8 | Fable 5.1 | xhigh | 1740 ±19 | Stirrup |
| 9 | Sonnet 5.5 | xhigh | 1735 ±23 | Stirrup |
| 10 | Gemini 4 Carbon | high | 1733 ±24 | Stirrup |
| 11 | Opus 5 | max | 1730 ±19 | Stirrup |
| 12 | Grok 4.7 | xhigh | 1718 ±21 | Stirrup |
| 13 | Grok 4.7 | high | 1716 ±19 | Stirrup |
| 14 | Opus 5.5 | high | 1714 ±21 | Stirrup |
| 15 | Opus 5 | xhigh | 1696 ±18 | Stirrup |
| 16 | MiMo V2.6 Pro | 1689 ±21 | Stirrup | |
| 17 | Muse Spark 1.3 | max | 1681 ±19 | Stirrup |
| 18 | GLM 5.3 | max | 1657 ±19 | Stirrup |
| 19 | Grok 4.6 | xhigh | 1654 ±19 | Stirrup |
| 20 | GLM 5.3 Flash | max | 1649 ±17 | Stirrup |
| 21 | Fable 5.1 | high | 1644 ±18 | Stirrup |
| 22 | Qwen 3.8 Flash | 1636 ±18 | Stirrup | |
| 23 | Muse Spark 1.3 | xhigh | 1632 ±18 | Stirrup |
| 24 | Haiku 5.5 | max | 1631 ±22 | Stirrup |
| 25 | Gemini 4 Argon | high | 1631 ±20 | Stirrup |
| 26 | Inkling 2 Small | 1630 ±21 | Stirrup | |
| 27 | Ling 3.1 Flash | 1626 ±21 | Stirrup | |
| 28 | Grok 4.6 | high | 1626 ±16 | Stirrup |
| 29 | Grok 4.6 | medium | 1620 ±18 | Stirrup |
| 30 | Fable 5 | max | 1618 ±16 | Stirrup |