Greek MMLU - Greek-Specific — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 62 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Preview) | 95.44 |
| 2 | GPT-4o (2024-11-20) | 93.11 |
| 3 | GPT-5.2 | 92.92 |
| 4 | Llama 3.1 70B Instruct | 86.99 |
| 5 | Llama 3.3 70B Instruct | 86.94 |
| 6 | Gemma 3 27B (IT) | 85.33 |
| 7 | Qwen 2.5 72B Instruct | 84.67 |
| 8 | Qwen 2.5 72B | 83.91 |
| 9 | gemma-3-27B-pt | 83.33 |
| 10 | Llama 3.1 70B | 82.46 |
| 11 | Gemma 3 12B (IT) | 82.21 |
| 12 | Qwen 3 30B A3B 2507 Instruct | 81.8 |
| 13 | gemma-3-12B-pt | 80.71 |
| 14 | Qwen 2.5 32B Instruct | 80.03 |
| 15 | EuroLLM-22B-Instruct-2512 | 78.99 |
Interactive version: aibenchmarks.dev/benchmark?slug=greek-mmlu-greek-specific · How the rankings work · Data refreshed daily, snapshot 2026-07-08.