all the models — AI benchmark observatory
← Benchmarks

GDPval-AA

GDPval-AA evaluates AI agents on economically valuable professional knowledge-work tasks and reports performance as an Elo score.

id gdpval-aa · max 3000 · 12 models reported

#ModelScore
1Claude Fable 5.1
Anthropic
1853.00
2GLM-5.3-Flash
Zhipu AI · open
1773.00
3GLM-5.3
Zhipu AI · open
1769.00
4Muse Spark 1.3
Meta
1754.00
5Grok 4.6
xAI
1753.00
6Hy4 preview
Tencent · open
1678.00
7Atria Dawn Preview
Shanghai AI Laboratory · open
1583.00
8Gemini 3.8 Flash
Google
1545.00
9Gemini 3.7 Flash
Google
1525.00
10Inkling-Small
Thinking Machines Lab · open
1269.00
11Inkling
Thinking Machines Lab · open
1238.00
12Ling 3.0 Flash
InclusionAI
1107.00