AA-Omniscience Accuracy: leaderboard
Artificial Analysis Omniscience accuracy on factual-recall questions across law, health, business, software engineering, humanities, and STEM.
Metric: Accuracy (%). Source: artificialanalysis.ai. Saturation forecast: Around July 2028. 580 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra (Max) | 62.6 | #6 (GPT-6 Astra) |
| 2 | GPT-6.1 Sol (Max) | 62.1 | #7 (GPT-6.1 Sol) |
| 3 | GPT-6 Astra (xHigh) | 61.85 | #6 (GPT-6 Astra) |
| 4 | GPT-6 Astra (High) | 61.13 | #6 (GPT-6 Astra) |
| 5 | Claude Opus 5 (Max) | 60.87 | #12 (Claude Opus 5) |
| 6 | GPT-6.1 Sol (xHigh) | 60.82 | #7 (GPT-6.1 Sol) |
| 7 | GPT-6.1 Sol (High) | 60.8 | #7 (GPT-6.1 Sol) |
| 8 | GPT-6 Astra (Medium) | 60.57 | #6 (GPT-6 Astra) |
| 9 | GPT-6.1 Sol (Medium) | 60.4 | #7 (GPT-6.1 Sol) |
| 10 | GPT-6 Astra (Low) | 59.53 | #6 (GPT-6 Astra) |
| 11 | Claude Opus 5 (xHigh) | 59.5 | #12 (Claude Opus 5) |
| 12 | GPT-5.6 Sol (Max) | 59.4 | #19 (GPT-5.6 Sol) |
| 13 | Claude Opus 5 (High) | 58.88 | #12 (Claude Opus 5) |
| 14 | GPT-6.1 Sol (Low) | 58.87 | #7 (GPT-6.1 Sol) |
| 15 | GPT-5.6 Sol (xHigh) | 58.82 | #19 (GPT-5.6 Sol) |
No result here: #3 Claude Opus 5.5, #9 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=aa-omniscience-accuracy · How It Works · Data refreshed daily, snapshot 2026-10-20.