AGI-Eval Community - Objective Accuracy (Chinese): 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 Chinese-language item split (_中) of the objective accuracy 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: Around February 2027. 139 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 84.95 | #33 |
| 2 | Claude Fable 5 | 83.67 | #20 |
| 3 | Gemini 3.1 Pro (Preview) | 82.1 | #64 |
| 4 | Kimi K3 | 81.91 | #35 |
| 5 | Gemini 3.5 Flash | 80.6 | #66 |
| 6 | Seed 2.1 Pro | 80.47 | #73 |
| 7 | GPT-5.4 | 79.68 | #83 |
| 8 | Qwen 3.7 Max | 79.26 | #85 |
| 9 | Claude Opus 4.8 | 79.25 | #39 |
| 10 | DeepSeek V4 Pro (Max) | 79.17 | #112 (DeepSeek V4 Pro) |
| 11 | Seed 2.0 Pro | 78.9 | #122 |
| 12 | Kimi K2.6 | 78.69 | #113 |
| 13 | Claude Sonnet 5 | 78.07 | #65 |
| 14 | Claude Opus 4.6 | 77.91 | #67 |
| 15 | GLM-5.2 | 77.86 | #80 |
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-objective-accuracy-chinese · How It Works · Data refreshed daily, snapshot 2026-11-05.