AGI-Eval Community Base - Objective Accuracy: leaderboard

AGI-Eval Community private-set Base Language Models evaluation (Objective Accuracy) 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 Base71.25#939
2DeepSeek-V3.1-Base68.22
3DeepSeek-V3.2-Exp-Base67.28
4DeepSeek V3 Base66.88
5ERNIE-4.5-300B-A47B-Base65.82
6Llama 4 Maverick Base65.29
7Qwen 2.5 32B (Base)62.95
8Qwen2.5-72B-Base62.87
9Qwen3-30B-A3B-Base62.85#1475
10Qwen3-14B-Base61.17#1473
11GLM-4.5-Base61.1
12Qwen3-8B-Base58.37#1754
13Llama 3.1 405B Base58.19
14Qwen2.5-14B-Base57.48
15Qwen3-4B-Base54.94#2439

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