GDPval-AA — leaderboard
Artificial Analysis implementation of GDPval, evaluating agent performance on economically valuable knowledge-work tasks across real occupations.
Metric: Elo. Source: artificialanalysis.ai. Status: saturated. 170 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.6 Sol (Max) | 1736 |
| 2 | GPT-5.6 Sol (xHigh) | 1702 |
| 3 | Kimi K3 | 1679 |
| 4 | GPT-5.6 Sol (High) | 1630 |
| 5 | Claude Sonnet 5 (Adaptive Reasoning, Max Effort) | 1607 |
| 6 | Claude Opus 4.8 (Adaptive Reasoning, Max Effort) | 1594 |
| 7 | GPT-5.6 Luna (Max) | 1584 |
| 8 | GPT-5.6 Terra (Max) | 1581 |
| 9 | GPT-5.6 Terra (xHigh) | 1572 |
| 10 | GPT-5.6 Sol (Medium) | 1562 |
| 11 | GPT-5.6 Luna (xHigh) | 1539 |
| 12 | Grok 4.5 (High) | 1535 |
| 13 | GLM-5.2 (Max) | 1514 |
| 14 | GPT-5.6 Terra (High) | 1513 |
| 15 | Claude Opus 4.7 (Adaptive Reasoning, Max Effort) | 1495 |
Interactive version: theaggregate.ai/benchmark?slug=gdpval-aa · How the rankings work · Data refreshed daily, snapshot 2026-07-22.