AA-Briefcase - Analytical Quality Elo: leaderboard
Artificial Analysis's private AA-Briefcase evaluation of long-horizon business workflows that produce spreadsheets, presentations and memos: Elo rating for the analytical quality of the deliverables. Higher is better; the scale is floored at 0 and several weak models are published at exactly 0.
Metric: Elo. Source: artificialanalysis.ai. Saturation forecast: Around January 2028. 249 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Muse Wonder Full (Max) | 2116.26 | #12 (Muse Wonder Full) |
| 2 | Kimi K3.1 Pro | 1982.53 | #17 |
| 3 | Muse Wonder (Max) | 1962.43 | #18 (Muse Wonder) |
| 4 | Qwen 3.8 Max (0902) | 1942.29 | #36 |
| 5 | Grok 4.7 (xHigh) | 1932.49 | #26 (Grok 4.7) |
| 6 | Grok 4.7 (High) | 1914.59 | #26 (Grok 4.7) |
| 7 | Claude Opus 5 (Max) | 1900.55 | #16 (Claude Opus 5) |
| 8 | Claude Opus 5 (xHigh) | 1894.67 | #16 (Claude Opus 5) |
| 9 | Claude Haiku 5.5 (Max) | 1893.66 | #43 (Claude Haiku 5.5) |
| 10 | Qwen3.8-Flash-Next | 1864.64 | #76 |
| 11 | Claude Opus 5 (High) | 1801.93 | #16 (Claude Opus 5) |
| 12 | Claude Haiku 5.5 (xHigh) | 1794.12 | #43 (Claude Haiku 5.5) |
| 13 | MiMo-V2.6-Pro | 1792.19 | #48 |
| 14 | DeepSeek V4.1 Flash (1017) (Max) | 1765.78 | #27 (DeepSeek V4.1 Flash (1017)) |
| 15 | GPT-6.1 Sol (Max) | 1762.52 | #7 (GPT-6.1 Sol) |
No result here: #3 Claude Opus 5.5, #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=aa-briefcase-analytical-quality-elo · How It Works · Data refreshed daily, snapshot 2026-11-05.