EuroEval Hungarian NLU - HuSST: leaderboard
The "HuSST" 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 | Llama 3.1 70B | 66.43 | #592 |
| 2 | Llama 3.3 70B Instruct | 65.96 | #534 |
| 3 | Gemma 3 27B (IT) | 65.07 | #523 |
| 4 | Qwen 2.5 72B Instruct | 63.65 | #454 |
| 5 | Qwen 3 30B A3B 2507 Instruct | 63.13 | #488 |
| 6 | Gemma 3 12B (IT) | 62.89 | #664 |
| 7 | Qwen 3 32B (Non-reasoning) | 62.53 | #443 (Qwen 3 32B) |
| 8 | Qwen 3 14B (Non-reasoning) | 62.4 | #537 (Qwen 3 14B) |
| 9 | Granite 4.0 H Small | 62.31 | #665 |
| 10 | gemma-3-27B-pt | 62.25 | #502 |
| 11 | gemma-3-12B-pt | 61.52 | #686 |
| 12 | EuroLLM-9B | 61.34 | #1254 |
| 13 | Mistral Small 3.1 | 60.95 | #610 |
| 14 | Mistral Small 3.2 | 60.01 | #605 |
| 15 | Gemma 3n E4B (IT) | 60 | #749 |
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-husst · How It Works · Data refreshed daily, snapshot 2026-10-20.