EuroEval Hungarian NLU - HuSST — leaderboard
Metric: Sentiment classification Score (%). Source: euroeval.com. 203 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.1 70B | 66.43 |
| 2 | Llama 3.3 70B Instruct | 65.96 |
| 3 | Gemma 3 27B (IT) | 65.07 |
| 4 | Qwen 2.5 72B Instruct | 63.65 |
| 5 | Qwen 3 30B A3B 2507 Instruct | 63.13 |
| 6 | Gemma 3 12B (IT) | 62.89 |
| 7 | Qwen 3 14B (Non-reasoning) | 62.4 |
| 8 | Granite 4.0 H Small | 62.31 |
| 9 | gemma-3-27B-pt | 62.25 |
| 10 | gemma-3-12B-pt | 61.52 |
| 11 | Mistral Small 3.1 | 60.95 |
| 12 | Mistral Small 3.2 | 60.01 |
| 13 | Gemma 3n E4B (IT) | 60 |
| 14 | Gemma 4 31B (IT) | 59.56 |
| 15 | Llama 3.1 8B Instruct | 58.8 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-hungarian-nlu-husst · How the rankings work · Data refreshed daily, snapshot 2026-07-22.