EuroEval Slovene NLU - SSJ500K NER: leaderboard
Metric: Named entity recognition Score (%). Source: euroeval.com. 204 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 Mini (Medium) | 83.51 |
| 2 | GPT-5.4 Mini (High) | 80.44 |
| 3 | Claude Sonnet 4.5 (Thinking) | 76.38 |
| 4 | GPT-5 Nano | 75.39 |
| 5 | GPT-5 | 75.31 |
| 6 | Claude Sonnet 4.5 (Non-reasoning) | 74.26 |
| 7 | Qwen 3 4B 2507 (Thinking) | 74.2 |
| 8 | Gemma 4 31B (IT) | 73.49 |
| 9 | Gemma 4 26B A4B (IT) | 72.7 |
| 10 | OLMo 3.1 32B (Thinking) | 71.59 |
| 11 | Gemini 2.5 Flash (Thinking) | 71.5 |
| 12 | Grok 4.1 Fast (Reasoning) | 71.13 |
| 13 | Gemini 3 Pro (Preview) | 70.92 |
| 14 | Gemini 3.1 Flash Lite (Preview) | 70.24 |
| 15 | Claude Sonnet 4.6 | 69.54 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-slovene-nlu-ssj500k-ner · How It Works · Data refreshed daily, snapshot 2026-09-05.