AA GDPval: leaderboard
Artificial Analysis independent evaluation of GDPval: real-world agentic tasks across 44 occupations and 9 industries with shell/web access. ELO-rated, 220 tasks.
Metric: ELO. Source: artificialanalysis.ai. Saturation forecast: Around January 2028. 298 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Muse Wonder Full (Max) | 1857.09 | #10 (Muse Wonder Full) |
| 2 | Muse Wonder (Max) | 1770.34 | #15 (Muse Wonder) |
| 3 | DeepSeek V4.1 Flash (1017) (Max) | 1745.77 | #20 (DeepSeek V4.1 Flash (1017)) |
| 4 | Claude Opus 5 (Max) | 1721.15 | #12 (Claude Opus 5) |
| 5 | Grok 4.7 (xHigh) | 1711.07 | #25 (Grok 4.7) |
| 6 | Grok 4.7 (High) | 1706.28 | #25 (Grok 4.7) |
| 7 | Claude Opus 5 (xHigh) | 1690.67 | #12 (Claude Opus 5) |
| 8 | MiMo-V2.6-Pro | 1684.34 | #42 |
| 9 | Muse Spark 1.3 (Max) | 1680.99 | #24 (Muse Spark 1.3) |
| 10 | Qwen 3.8 Max (0902) | 1665.58 | #29 |
| 11 | GLM-5.3 (Max) | 1650.31 | #39 (GLM-5.3) |
| 12 | Grok 4.6 (xHigh) | 1645.38 | #26 (Grok 4.6) |
| 13 | GLM-5.3 Flash | 1642.18 | #65 |
| 14 | Qwen3.8-Flash-Next | 1627.86 | #72 |
| 15 | Muse Spark 1.3 (xHigh) | 1627.25 | #24 (Muse Spark 1.3) |
No result here: #3 Claude Opus 5.5, #9 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=aa-gdpval · How It Works · Data refreshed daily, snapshot 2026-10-20.