AGI-Eval Community - Cognition (English): leaderboard
Metric: Accuracy (%). Source: agi-eval.cn. 139 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4 (Thinking) | 89.88 |
| 2 | Claude Opus 4.6 | 88.41 |
| 3 | Grok 4 | 87.63 |
| 4 | Claude Opus 4.5 (Non-reasoning) | 87.19 |
| 5 | Gemini 3 Flash (Preview) | 87.06 |
| 6 | Gemini 3.1 Pro (Preview) | 86.97 |
| 7 | Gemini 3.5 Flash | 86.85 |
| 8 | Claude Fable 5 | 86.78 |
| 9 | Seed 1.8 | 86.63 |
| 10 | Claude Sonnet 4 (Thinking) | 86.34 |
| 11 | Seed 2.0 Pro | 86.05 |
| 12 | Kimi K2.6 | 85.74 |
| 13 | Claude Opus 4.8 | 85.69 |
| 14 | O3 Pro | 85.56 |
| 15 | Doubao-Seed-1.6 (Thinking) | 85.56 |
Interactive version: theaggregate.ai/benchmark?slug=agi-eval-community-cognition-english · How It Works · Data refreshed daily, snapshot 2026-09-19.