Greek MMLU - Humanities: leaderboard
The Humanities subtask of Greek MMLU — humanities questions in Greek; accuracy (%).
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 84 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash (Preview) | 92.88 | #82 |
| 2 | GPT-4o (2024-11-20) | 88.68 | #381 |
| 3 | GPT-5.2 | 88.27 | #110 |
| 4 | Llama 3.1 70B Instruct | 82.34 | #561 |
| 5 | Llama 3.3 70B Instruct | 82.2 | #534 |
| 6 | Gemma 3 27B (IT) | 79.87 | #523 |
| 7 | gemma-3-27B-pt | 79.83 | #502 |
| 8 | Qwen 2.5 72B Instruct | 79.14 | #454 |
| 9 | Llama 3.1 70B | 79.05 | #592 |
| 10 | Qwen 2.5 72B | 78.78 | #499 |
| 11 | EuroLLM-22B-Instruct-2512 | 75.31 | #701 |
| 12 | Gemma 3 12B (IT) | 75.13 | #664 |
| 13 | Qwen 3 30B A3B 2507 Instruct | 74.81 | #488 |
| 14 | gemma-3-12B-pt | 74.44 | #686 |
| 15 | Mistral-Small-24B-Base-2501 | 74.07 | #670 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=greek-mmlu-humanities · How It Works · Data refreshed daily, snapshot 2026-10-20.