AA-Briefcase: leaderboard
Artificial Analysis Briefcase: agentic knowledge-work tasks over a briefcase of real-world files; scored by rubric pass rate, the cross-model-comparable metric behind the site's quality Elo.
Metric: Rubric Pass Rate (%). Source: artificialanalysis.ai. Saturation forecast: Around June 2027. 233 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon (High) | 64.63 | #8 (Gemini 4 Argon) |
| 2 | Muse Wonder Full (Max) | 61.31 | #10 (Muse Wonder Full) |
| 3 | Muse Wonder (Max) | 61.21 | #15 (Muse Wonder) |
| 4 | Muse Spark 1.3 (Max) | 58.69 | #24 (Muse Spark 1.3) |
| 5 | Claude Opus 5 (xHigh) | 57.98 | #12 (Claude Opus 5) |
| 6 | Qwen 3.8 Max (0902) | 57.96 | #29 |
| 7 | Claude Opus 5 (Max) | 57.17 | #12 (Claude Opus 5) |
| 8 | GPT-6.1 Sol (Max) | 55.66 | #7 (GPT-6.1 Sol) |
| 9 | Grok 4.7 (xHigh) | 55.35 | #25 (Grok 4.7) |
| 10 | Grok 4.7 (High) | 55.01 | #25 (Grok 4.7) |
| 11 | MiMo-V2.6-Pro | 54.75 | #42 |
| 12 | Qwen3.8-Flash-Next | 54.1 | #72 |
| 13 | Claude Opus 5 (High) | 53.84 | #12 (Claude Opus 5) |
| 14 | Grok 4.6 (xHigh) | 53.84 | #26 (Grok 4.6) |
| 15 | Grok 4.6 (High) | 52.83 | #26 (Grok 4.6) |
No result here: #3 Claude Opus 5.5, #9 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=aa-briefcase · How It Works · Data refreshed daily, snapshot 2026-10-20.