EuroEval Croatian NLU - WikiANN HR: leaderboard
Metric: Named entity recognition Score (%). Source: euroeval.com. 211 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 (Thinking) | 69.26 |
| 2 | Gemini 3 Pro (Preview) | 68.98 |
| 3 | Gemma 4 31B (IT) | 68.87 |
| 4 | Mistral Small 3.2 | 68.75 |
| 5 | Qwen 3 4B 2507 (Thinking) | 67.93 |
| 6 | GPT-5.4 Mini (High) | 67.85 |
| 7 | Mistral Small 3.1 | 67.69 |
| 8 | OLMo 3.1 32B (Thinking) | 67.54 |
| 9 | Llama 3.1 70B | 67.53 |
| 10 | MamayLM-Gemma-3-12B-IT-v1.0 | 67.14 |
| 11 | Gemma 4 26B A4B (IT) | 66.93 |
| 12 | gemma-3-27B-pt | 66.66 |
| 13 | GPT-OSS-20B (Low) | 66.58 |
| 14 | Grok 4.20 Beta (0309) (Reasoning) | 66.45 |
| 15 | GPT-5.4 Mini (Medium) | 66.38 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-croatian-nlu-wikiann-hr · How It Works · Data refreshed daily, snapshot 2026-09-05.