AA-AnalystAgent - Pass@1: leaderboard
Artificial Analysis evaluates agents on data-analysis work. The published headline score, pass-at-one and pass-at-k rates are retained as separate measurements of task success.
Metric: Pass@1 (%). Source: artificialanalysis.ai. Saturation forecast: Around February 2028. 51 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.7 Flash (High) | 70.5 | #24 (Gemini 3.7 Flash) |
| 2 | Claude Haiku 5.5 (Max) | 68.25 | #43 (Claude Haiku 5.5) |
| 3 | Grok 4.7 (High) | 67 | #26 (Grok 4.7) |
| 4 | GPT-5.5 (xHigh) | 66.25 | #33 (GPT-5.5) |
| 5 | GPT-6 Luna (Max) | 65.75 | #61 (GPT-6 Luna) |
| 6 | Gemini 3.1 Pro (Preview) | 64 | #64 |
| 7 | Claude Opus 5 (Max) | 63.75 | #16 (Claude Opus 5) |
| 8 | Grok 4.7 (xHigh) | 63.5 | #26 (Grok 4.7) |
| 9 | Claude Opus 4.8 (Max) | 63.5 | #39 (Claude Opus 4.8) |
| 10 | Claude Sonnet 5 (Max) | 61.5 | #65 (Claude Sonnet 5) |
| 11 | GPT-5.6 Sol (Max) | 61.25 | #22 (GPT-5.6 Sol) |
| 12 | Gemini 3.8 Flash (High) | 61.25 | #30 (Gemini 3.8 Flash) |
| 13 | MiMo-V2.6-Pro | 61.25 | #48 |
| 14 | Qwen 3.8 Max (0902) | 61.25 | #36 |
| 15 | Grok 4.6 (High) | 60.75 | #29 (Grok 4.6) |
No result here: #3 Claude Opus 5.5, #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=aa-analystagent-pass-1 · How It Works · Data refreshed daily, snapshot 2026-11-05.