AGI-Eval Community Base - Interaction: leaderboard
AGI-Eval Community private-set Base Language Models evaluation (Interaction) measuring accuracy (%) across bilingual capability dimensions.
Metric: Accuracy (%). Source: agi-eval.org. Saturation forecast: Estimated already saturated. 25 models tracked. Updated 22 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen2.5-72B-Base | 81.21 | |
| 2 | Qwen 2.5 32B (Base) | 80.78 | |
| 3 | DeepSeek-V3.1-Base | 79.87 | |
| 4 | Kimi K2 Base | 79.83 | #939 |
| 5 | DeepSeek V3 Base | 79.45 | |
| 6 | DeepSeek-V3.2-Exp-Base | 78.38 | |
| 7 | ERNIE-4.5-300B-A47B-Base | 78.3 | |
| 8 | Qwen2.5-14B-Base | 78.16 | |
| 9 | Llama 4 Maverick Base | 77.13 | |
| 10 | Llama 3.1 405B Base | 77.12 | |
| 11 | Qwen3-14B-Base | 76.91 | #1473 |
| 12 | Qwen3-8B-Base | 76.86 | #1754 |
| 13 | GLM4-9B-Base | 76.75 | |
| 14 | GLM-4.5-Base | 75.96 | |
| 15 | Llama 3.1 70B Base | 75.04 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=agi-eval-community-base-interaction · How It Works · Data refreshed daily, snapshot 2026-11-22.