EuroEval Estonian NLU - Grammar ET — leaderboard
Metric: Linguistic acceptability Score (%). Source: euroeval.com. 250 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Preview) (Non-reasoning) | 66.32 |
| 2 | Gemini 3 Pro (Preview) | 62.63 |
| 3 | Gemini 2.5 Pro | 61.06 |
| 4 | Gemini 3.1 Flash Lite (Preview) | 58.7 |
| 5 | Gemini 2.5 Flash (Thinking) | 56 |
| 6 | GPT-5 | 55.2 |
| 7 | GPT-5.4 Mini (High) | 53.72 |
| 8 | Claude Sonnet 4.6 | 52.96 |
| 9 | GPT-4.1 | 52.89 |
| 10 | EuroLLM-22B-Instruct-2512 | 51.64 |
| 11 | GPT-5.4 Mini (Medium) | 49.33 |
| 12 | GPT-5.4 Nano (Medium) | 48.69 |
| 13 | GPT-5.4 Mini (Non-reasoning) | 48.19 |
| 14 | GPT-5.2 | 46.89 |
| 15 | GPT-5 Mini | 46.35 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-estonian-nlu-grammar-et · How the rankings work · Data refreshed daily, snapshot 2026-07-22.