EuroEval Hungarian NLU - ScaLA HU: leaderboard
The "ScaLA HU" subtask of EuroEval Hungarian NLU. Hungarian-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: Matthews correlation coefficient (%). Source: euroeval.com. Saturation forecast: Around January 2027. 203 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 79.09 | #66 |
| 2 | Gemini 3 Flash (Preview) (Non-reasoning) | 74.15 | #82 (Gemini 3 Flash (Preview)) |
| 3 | Gemini 3 Flash (Preview) (Thinking) | 73.42 | #82 (Gemini 3 Flash (Preview)) |
| 4 | Gemini 2.5 Pro | 72.72 | #152 |
| 5 | GPT-4.1 | 70.5 | #248 |
| 6 | Gemini 3.1 Flash Lite (Preview) | 69.59 | #194 |
| 7 | Claude Sonnet 4.5 (Thinking) | 68.18 | #142 (Claude Sonnet 4.5) |
| 8 | GPT-5.4 Mini (2026-03-17) (High) | 67.89 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 9 | Gemini 2.5 Flash (Thinking) | 67.25 | #245 (Gemini 2.5 Flash) |
| 10 | GPT-5.2 | 65.68 | #110 |
| 11 | GPT-5 | 64 | #98 |
| 12 | Claude Sonnet 4.6 | 63.91 | #90 |
| 13 | GPT-5 Mini (2025-08-07) | 60.51 | #172 |
| 14 | Gemma 4 31B (IT) | 59.7 | #176 |
| 15 | Gemini 2.5 Flash Lite (Thinking) | 58.78 | #422 (Gemini 2.5 Flash Lite) |
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-scala-hu · How It Works · Data refreshed daily, snapshot 2026-10-20.