MT-Bench PL - STEM — leaderboard
Metric: Judge Score (0-10). Source: huggingface.co. 50 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 3 12B (IT) | 10 |
| 2 | Phi-4 | 10 |
| 3 | Gemma 3 27B (IT) | 9.95 |
| 4 | aya-expanse-32B | 9.95 |
| 5 | Mistral Small 3.1 | 9.9 |
| 6 | Gemma 2 27B (IT) | 9.8 |
| 7 | aya-expanse-8B | 9.75 |
| 8 | Qwen 2.5 32B Instruct | 9.7 |
| 9 | Gemma 3 4B (IT) | 9.65 |
| 10 | Mistral-Small-Instruct-2409 | 9.65 |
| 11 | Qwen 2.5 14B Instruct | 9.6 |
| 12 | Llama 3.1 70B Instruct | 9.55 |
| 13 | Qwen 2 72B Instruct | 9.55 |
| 14 | Mistral Small 3 | 9.5 |
| 15 | Bielik-11B-v2.2-Instruct | 9.45 |
Interactive version: theaggregate.ai/benchmark?slug=mt-bench-pl-stem · How the rankings work · Data refreshed daily, snapshot 2026-07-22.