EuroEval Greek NLU - Greek SA — leaderboard
Metric: Sentiment classification Score (%). Source: euroeval.com. 204 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | gemma-3-12B-pt | 84 |
| 2 | Llama 3.3 70B Instruct | 83.74 |
| 3 | Llama 3.1 70B | 83.3 |
| 4 | Gemma 3 27B (IT) | 82.43 |
| 5 | Gemma 4 31B (IT) | 81.64 |
| 6 | gemma-3-27B-pt | 81.08 |
| 7 | Qwen 3 30B A3B 2507 Instruct | 80.73 |
| 8 | Grok 4.1 Fast (Reasoning) | 80.68 |
| 9 | Gemma 3 12B (IT) | 80.58 |
| 10 | GPT-4.1 | 80.5 |
| 11 | Claude Sonnet 4.5 (Thinking) | 79.99 |
| 12 | Claude Sonnet 4.6 | 79.79 |
| 13 | Qwen 2.5 72B Instruct | 79.41 |
| 14 | GPT-5.2 | 79.33 |
| 15 | Gemini 3 Flash (Preview) (Non-reasoning) | 79.07 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-greek-nlu-greek-sa · How the rankings work · Data refreshed daily, snapshot 2026-07-22.