EuroEval Lithuanian NLU - ScaLA LT: leaderboard
The "ScaLA LT" subtask of EuroEval Lithuanian NLU. Lithuanian-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 2028. 230 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 64.96 | #66 |
| 2 | Gemini 3 Flash (Preview) (Non-reasoning) | 60.29 | #82 (Gemini 3 Flash (Preview)) |
| 3 | Claude Sonnet 4.6 | 59.11 | #90 |
| 4 | Gemini 3 Flash (Preview) (Thinking) | 57.82 | #82 (Gemini 3 Flash (Preview)) |
| 5 | GPT-5 | 56.77 | #98 |
| 6 | GPT-5.4 Mini (2026-03-17) (High) | 53.61 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 7 | Gemini 3.1 Flash Lite (Preview) | 52.41 | #194 |
| 8 | Gemini 2.5 Pro | 51.54 | #152 |
| 9 | Claude 3.5 Haiku (20241022) | 50.62 | #588 |
| 10 | GPT-5 Mini (2025-08-07) | 49.84 | #172 |
| 11 | Qwen 3 235B A22B 2507 Instruct | 47.37 | #300 |
| 12 | GPT-4.1 | 46.15 | #248 |
| 13 | Gemini 2.5 Flash (Thinking) | 45.05 | #245 (Gemini 2.5 Flash) |
| 14 | GPT-5.2 | 42.68 | #110 |
| 15 | Gemma 3 27B (IT) | 41.68 | #523 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-lithuanian-nlu-scala-lt · How It Works · Data refreshed daily, snapshot 2026-10-20.