Greek MMLU - Greek-Specific: leaderboard
The Greek-Specific subtask of Greek MMLU — questions on Greek language, history and culture; 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) | 95.44 | #82 |
| 2 | GPT-4o (2024-11-20) | 93.11 | #381 |
| 3 | GPT-5.2 | 92.92 | #110 |
| 4 | Llama 3.1 70B Instruct | 86.99 | #561 |
| 5 | Llama 3.3 70B Instruct | 86.94 | #534 |
| 6 | Gemma 3 27B (IT) | 85.33 | #523 |
| 7 | Qwen 2.5 72B Instruct | 84.67 | #454 |
| 8 | Qwen 2.5 72B | 83.91 | #499 |
| 9 | gemma-3-27B-pt | 83.33 | #502 |
| 10 | Llama 3.1 70B | 82.46 | #592 |
| 11 | Gemma 3 12B (IT) | 82.21 | #664 |
| 12 | Qwen 3 30B A3B 2507 Instruct | 81.8 | #488 |
| 13 | gemma-3-12B-pt | 80.71 | #686 |
| 14 | Qwen 2.5 32B Instruct | 80.03 | #498 |
| 15 | EuroLLM-22B-Instruct-2512 | 78.99 | #701 |
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-greek-specific · How It Works · Data refreshed daily, snapshot 2026-10-20.