TMMLU+ - Humanities: leaderboard
The "Humanities" category of TMMLU+, the Traditional Chinese multi-subject exam benchmark (ikala): 66 subjects drawn from Taiwanese professional, high-school and university examinations, answered as multiple choice and reported per category as well as overall.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Around February 2027. 21 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 | 94.6 | #16 |
| 2 | Claude Fable 5 | 90.02 | #20 |
| 3 | Gemini 3.1 Pro (Preview) (Thinking) | 84.91 | #64 (Gemini 3.1 Pro (Preview)) |
| 4 | GPT-5.6 Sol | 82.3 | #22 |
| 5 | Claude Sonnet 5 | 81.72 | #65 |
| 6 | DeepSeek V4 Pro (0813) | 79.63 | #58 |
| 7 | DeepSeek V4 Flash | 77.42 | #155 |
| 8 | Kimi K3 | 77.31 | #35 |
| 9 | Qwen 3.7 Max | 76.03 | #85 |
| 10 | Hy3 | 74.75 | #100 |
| 11 | GPT-5.6 Terra | 72.49 | #40 |
| 12 | GPT-5.6 Luna | 70.81 | #79 |
| 13 | GLM-5.2 | 70.28 | #80 |
| 14 | Grok 4.3 | 67.67 | #204 |
| 15 | MiMo-V2.5 | 67.32 | #196 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=tmmlu-plus-humanities · How It Works · Data refreshed daily, snapshot 2026-11-05.