MT-Bench PL - Math — leaderboard
Metric: Judge Score (0-10). Source: huggingface.co. 50 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 3 27B (IT) | 8.25 |
| 2 | Qwen 2.5 14B Instruct | 8.1 |
| 3 | Mistral Small 3.1 | 7.85 |
| 4 | Mistral Small 3 | 7.83 |
| 5 | Gemma 2 27B (IT) | 7.8 |
| 6 | Mistral Large 2 (Jul) | 7.8 |
| 7 | Phi-4 | 7.7 |
| 8 | Bielik-11B-v2.3-Instruct | 7.7 |
| 9 | Qwen 2.5 32B Instruct | 7.6 |
| 10 | Gemma 3 12B (IT) | 7.45 |
| 11 | Gemma 3 4B (IT) | 7.4 |
| 12 | Mistral-Small-Instruct-2409 | 7 |
| 13 | Mixtral 8x22B | 6.9 |
| 14 | GPT-3.5 Turbo | 6.85 |
| 15 | Mistral Nemo Instruct (2407) | 6.7 |
Interactive version: theaggregate.ai/benchmark?slug=mt-bench-pl-math · How the rankings work · Data refreshed daily, snapshot 2026-07-22.