Greek MMLU - STEM — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 62 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Preview) | 92.82 |
| 2 | GPT-5.2 | 86.05 |
| 3 | GPT-4o (2024-11-20) | 84.54 |
| 4 | Qwen 3 30B A3B 2507 Instruct | 79.31 |
| 5 | Qwen 2.5 72B Instruct | 78.9 |
| 6 | Qwen 2.5 72B | 78.31 |
| 7 | Gemma 3 27B (IT) | 78.03 |
| 8 | gemma-3-27B-pt | 77.81 |
| 9 | Llama 3.3 70B Instruct | 77.03 |
| 10 | Llama 3.1 70B Instruct | 76.35 |
| 11 | gemma-3-12B-pt | 73.3 |
| 12 | Gemma 3 12B (IT) | 72.95 |
| 13 | Qwen 2.5 32B | 72.74 |
| 14 | Llama 3.1 70B | 72.42 |
| 15 | Qwen 2.5 32B Instruct | 72.08 |
Interactive version: aibenchmarks.dev/benchmark?slug=greek-mmlu-stem · How the rankings work · Data refreshed daily, snapshot 2026-07-08.