EuroEval Polish NLU - KPWr NER: leaderboard
Metric: Named entity recognition Score (%). Source: euroeval.com. 234 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 Mini (High) | 78.64 |
| 2 | GPT-5 | 77.08 |
| 3 | GPT-5.4 Mini (Medium) | 75.67 |
| 4 | Claude Sonnet 4.5 (Thinking) | 74.13 |
| 5 | Gemma 4 31B (IT) | 72.58 |
| 6 | Gemma 4 26B A4B (IT) | 70.61 |
| 7 | Gemini 3 Pro (Preview) | 69.79 |
| 8 | Qwen 3 4B 2507 (Thinking) | 69.57 |
| 9 | Qwen 3 Next 80B A3B (Thinking) | 68.95 |
| 10 | Claude Sonnet 4.6 | 67.43 |
| 11 | Gemini 3.1 Flash Lite (Preview) | 67.37 |
| 12 | Grok 4.1 Fast (Reasoning) | 66.9 |
| 13 | Gemini 2.5 Pro | 66.29 |
| 14 | Gemini 2.5 Flash (Thinking) | 65.71 |
| 15 | Claude Sonnet 4.5 (Non-reasoning) | 65.55 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-polish-nlu-kpwr-ner · How It Works · Data refreshed daily, snapshot 2026-09-05.