MT-Bench PL - Reasoning: leaderboard
Metric: Judge Score (0-10). Source: huggingface.co. 50 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Phi-4 | 9.55 |
| 2 | Qwen 2.5 32B Instruct | 9.1 |
| 3 | Mistral Small 3.1 | 9 |
| 4 | aya-expanse-32B | 8.95 |
| 5 | Qwen 2 72B Instruct | 8.85 |
| 6 | Mistral Large 2 (Jul) | 8.7 |
| 7 | Gemma 3 27B (IT) | 8.4 |
| 8 | Mistral Small 3 | 7.9 |
| 9 | Mistral-Small-Instruct-2409 | 7.9 |
| 10 | Gemma 3 12B (IT) | 7.75 |
| 11 | Qwen 2.5 14B Instruct | 7.55 |
| 12 | Bielik-11B-v2.2-Instruct | 6.9 |
| 13 | Gemma 2 27B (IT) | 6.85 |
| 14 | aya-expanse-8B | 6.85 |
| 15 | Mixtral 8x22B | 6.3 |
Interactive version: theaggregate.ai/benchmark?slug=mt-bench-pl-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.