EuroEval Hungarian NLU: leaderboard
Hungarian-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: NLU Average Score (%). Source: euroeval.com. Saturation forecast: Estimated already saturated. 208 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 3.1 70B | 65.84 | #592 |
| 2 | Gemma 4 31B (IT) | 65.62 | #176 |
| 3 | Gemini 3 Pro (Preview) | 65.16 | #66 |
| 4 | Gemini 3 Flash (Preview) (Thinking) | 64.28 | #82 (Gemini 3 Flash (Preview)) |
| 5 | Gemini 3 Flash (Preview) (Non-reasoning) | 64 | #82 (Gemini 3 Flash (Preview)) |
| 6 | Gemma 3 27B (IT) | 63.45 | #523 |
| 7 | Claude Sonnet 4.5 (Thinking) | 63.03 | #142 (Claude Sonnet 4.5) |
| 8 | Gemini 2.5 Pro | 62.09 | #152 |
| 9 | Gemini 3.1 Flash Lite (Preview) | 61.85 | #194 |
| 10 | GPT-5.4 Mini (2026-03-17) (High) | 61.75 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 11 | gemma-3-27B-pt | 61.68 | #502 |
| 12 | Claude Sonnet 4.6 | 61.41 | #90 |
| 13 | Mistral Small 3.2 | 61.16 | #605 |
| 14 | Gemini 2.5 Flash (Thinking) | 60.96 | #245 (Gemini 2.5 Flash) |
| 15 | Mistral Small 3.1 | 60.42 | #610 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-hungarian-nlu · How It Works · Data refreshed daily, snapshot 2026-10-20.