EuroEval Polish NLU - KPWr NER: leaderboard
The "KPWr NER" subtask of EuroEval Polish NLU. Polish-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: Micro F1 excluding MISC (%). Source: euroeval.com. Saturation forecast: Around January 2027. 234 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 Mini (2026-03-17) (High) | 78.64 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 2 | GPT-5 | 77.08 | #98 |
| 3 | Claude Sonnet 4.5 (Thinking) | 74.13 | #142 (Claude Sonnet 4.5) |
| 4 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 72.87 | #246 (Qwen 3 235B A22B 2507 FP8) |
| 5 | Gemma 4 31B (IT) | 72.58 | #176 |
| 6 | Gemini 3 Flash (Preview) (Non-reasoning) | 71.92 | #82 (Gemini 3 Flash (Preview)) |
| 7 | Gemma 4 26B A4B (IT) | 70.61 | #297 |
| 8 | Gemini 3 Pro (Preview) | 69.79 | #66 |
| 9 | Qwen 3 4B 2507 (Thinking) | 69.57 | #538 (Qwen 3 4B 2507) |
| 10 | Qwen 3 Next 80B A3B (Thinking) | 68.95 | #320 (Qwen 3 Next 80B A3B) |
| 11 | Gemini 3 Flash (Preview) (Thinking) | 68.94 | #82 (Gemini 3 Flash (Preview)) |
| 12 | NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 | 68.86 | #679 |
| 13 | Qwen 3.5 35B A3B Base | 68.72 | #310 |
| 14 | Claude Sonnet 4.6 | 67.43 | #90 |
| 15 | Gemini 3.1 Flash Lite (Preview) | 67.37 | #194 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-polish-nlu-kpwr-ner · How It Works · Data refreshed daily, snapshot 2026-10-20.