AGI-Eval Community - Interaction (English): leaderboard
AGI-Eval community leaderboard (agi-eval.cn), the private-set LLM board built from the platform's non-public items; latest monthly snapshot. Unrelated to the AGIEval exam dataset. This board is the English-language item split (_英) of the interaction column; the parent column combines both splits. Upstream accuracy 0-1, shown as 0-100; higher is better.
Metric: Accuracy (%). Source: agi-eval.cn. Saturation forecast: Estimated already saturated. 139 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 95.45 | #64 |
| 2 | Gemini 3.5 Flash | 95.2 | #66 |
| 3 | Seed 2.1 Pro | 94.97 | #73 |
| 4 | Claude Fable 5 | 94.92 | #20 |
| 5 | Claude Opus 4.8 | 94.77 | #39 |
| 6 | Claude Opus 4.6 | 94.75 | #67 |
| 7 | Qwen 3.7 Max | 94.73 | #85 |
| 8 | Kimi K3 | 94.54 | #35 |
| 9 | Seed 2.0 Pro | 94.52 | #122 |
| 10 | GLM-5.2 | 94.38 | #80 |
| 11 | Gemini 3 Pro (Preview) | 94.31 | #71 |
| 12 | GPT-5.5 | 94.17 | #33 |
| 13 | Kimi K2.5 (Thinking) | 93.89 | #152 (Kimi K2.5) |
| 14 | Claude Sonnet 4.5 (Thinking) | 93.86 | #148 (Claude Sonnet 4.5) |
| 15 | Qwen 3.5 397B A17B (Thinking) | 93.84 | #154 (Qwen 3.5 397B A17B) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=agi-eval-community-interaction-english · How It Works · Data refreshed daily, snapshot 2026-11-05.