AA-Omniscience v1.1 Accuracy: leaderboard
Artificial Analysis's AA-Omniscience v1.1 accuracy: the share of its 3,000 factual-recall questions (Business, Health, Humanities & Social Sciences, Law, Software Engineering, Science, Engineering & Mathematics) answered correctly, 0-100 (%), higher is better; an abstention counts as not correct. Not directly comparable with v1's 6,000-question AA-Omniscience Accuracy.
Metric: Accuracy (%). Source: artificialanalysis.ai. Saturation forecast: Around April 2028. 134 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6.1 Sol (Max) | 61.83 | #10 (GPT-6.1 Sol) |
| 2 | GPT-6 Astra (Max) | 61.57 | #6 (GPT-6 Astra) |
| 3 | GPT-6 Astra (High) | 61.1 | #6 (GPT-6 Astra) |
| 4 | GPT-6.1 Sol (xHigh) | 61.07 | #10 (GPT-6.1 Sol) |
| 5 | GPT-6 Astra (xHigh) | 61.03 | #6 (GPT-6 Astra) |
| 6 | Claude Opus 5 (Max) | 60.87 | #20 (Claude Opus 5) |
| 7 | GPT-6.1 Sol (High) | 60.4 | #10 (GPT-6.1 Sol) |
| 8 | GPT-6 Astra (Medium) | 59.63 | #6 (GPT-6 Astra) |
| 9 | GPT-6.1 Sol (Medium) | 59.6 | #10 (GPT-6.1 Sol) |
| 10 | GPT-6.1 Sol (Low) | 58.63 | #10 (GPT-6.1 Sol) |
| 11 | GPT-6 Astra (Low) | 58.6 | #6 (GPT-6 Astra) |
| 12 | Gemini 4 Carbon (High) | 58.27 | #2 (Gemini 4 Carbon) |
| 13 | GPT-6 Sol (Max) | 51.13 | #24 (GPT-6 Sol) |
| 14 | Gemini 3.8 Flash (High) | 50.53 | #36 (Gemini 3.8 Flash) |
| 15 | hy4 | 49.97 | #40 |
No result here: #1 Claude Fable 5.2, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=aa-omniscience-v1-1-accuracy · How It Works · Data refreshed daily, snapshot 2026-11-12.