AGI-Eval Community - Interaction (English): leaderboard
Metric: Accuracy (%). Source: agi-eval.cn. 139 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 95.45 |
| 2 | Gemini 3.5 Flash | 95.2 |
| 3 | Seed 2.1 Pro | 94.97 |
| 4 | Claude Fable 5 | 94.92 |
| 5 | Claude Opus 4.8 | 94.77 |
| 6 | Claude Opus 4.6 | 94.75 |
| 7 | Qwen 3.7 Max | 94.73 |
| 8 | Kimi K3 | 94.54 |
| 9 | Seed 2.0 Pro | 94.52 |
| 10 | GLM-5.2 | 94.38 |
| 11 | Gemini 3 Pro (Preview) | 94.31 |
| 12 | GPT-5.5 | 94.17 |
| 13 | Kimi K2.5 (Thinking) | 93.89 |
| 14 | Claude Sonnet 4.5 (Thinking) | 93.86 |
| 15 | Qwen 3.5 397B A17B (Thinking) | 93.84 |
Interactive version: theaggregate.ai/benchmark?slug=agi-eval-community-interaction-english · How It Works · Data refreshed daily, snapshot 2026-09-19.