AGI-Eval Community Base - General Reasoning: leaderboard
AGI-Eval Community private-set Base Language Models evaluation (General Reasoning) 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 | ERNIE-4.5-300B-A47B-Base | 87.97 | |
| 2 | Kimi K2 Base | 86.66 | #939 |
| 3 | DeepSeek-V3.1-Base | 85.91 | |
| 4 | DeepSeek V3 Base | 85.42 | |
| 5 | GLM-4.5-Base | 84.94 | |
| 6 | DeepSeek-V3.2-Exp-Base | 84 | |
| 7 | Llama 4 Maverick Base | 82.6 | |
| 8 | Qwen2.5-72B-Base | 82.37 | |
| 9 | Llama 3.1 405B Base | 82.3 | |
| 10 | Qwen 2.5 32B (Base) | 81.33 | |
| 11 | Qwen3-14B-Base | 81.19 | #1473 |
| 12 | Qwen3-30B-A3B-Base | 77.55 | #1475 |
| 13 | Qwen2.5-14B-Base | 76.25 | |
| 14 | Qwen3-8B-Base | 75.21 | #1754 |
| 15 | Llama 3.1 70B Base | 74.63 |
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-general-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-22.