AGI-Eval Community Base - Objective Accuracy (English): leaderboard

AGI-Eval Community private-set Base Language Models evaluation (Objective Accuracy (English)) 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

#ModelScoreOverall rank
1Kimi K2 Base65.61#939
2DeepSeek-V3.1-Base64.61
3DeepSeek V3 Base63.77
4DeepSeek-V3.2-Exp-Base63.45
5Llama 4 Maverick Base62.24
6Qwen3-30B-A3B-Base60.86#1475
7Qwen3-14B-Base60.02#1473
8Qwen2.5-72B-Base59.52
9Qwen 2.5 32B (Base)59.3
10ERNIE-4.5-300B-A47B-Base58.98
11Qwen3-8B-Base56.29#1754
12Llama 3.1 405B Base56.17
13Qwen2.5-14B-Base54.76
14Qwen3-4B-Base52.52#2439
15GLM-4.5-Base52.25

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-objective-accuracy-english · How It Works · Data refreshed daily, snapshot 2026-11-22.