EuroEval Polish NLU - Polemo2: leaderboard
Metric: Sentiment classification Score (%). Source: euroeval.com. 234 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Ministral 3 3B | 98.75 |
| 2 | Ministral 3 14B | 98.2 |
| 3 | Llama 3.1 70B | 97.77 |
| 4 | Ministral 3 8B | 97.11 |
| 5 | Gemma 4 31B (IT) | 96.92 |
| 6 | GLM-4.7 Flash | 95.48 |
| 7 | Mistral Small 3.2 | 95.17 |
| 8 | Mistral Small 3.1 | 94.91 |
| 9 | Llama 3.1 8B | 94.7 |
| 10 | gemma-3-27B-pt | 94.61 |
| 11 | Magistral Small 1.2 | 94.56 |
| 12 | Llama 3.3 70B Instruct | 93.31 |
| 13 | Ministral-3-14B-Instruct-2512 | 93.21 |
| 14 | gemma-3-4B-pt Base | 92.84 |
| 15 | Qwen 3 30B A3B (Non-reasoning) | 92.12 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-polish-nlu-polemo2 · How It Works · Data refreshed daily, snapshot 2026-09-05.