EuroEval Lithuanian NLU - MultiWikiQA LT: leaderboard
The "MultiWikiQA 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: Character F1 (%). Source: euroeval.com. Saturation forecast: Around January 2027. 230 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 3.1 70B | 75.58 | #592 |
| 2 | gemma-3-27B-pt | 74.07 | #502 |
| 3 | Magistral Small 1.2 | 73.58 | #570 |
| 4 | Mistral Small 3.2 | 73.45 | #605 |
| 5 | Ministral-3-14B-Reasoning-2512 | 72.63 | #540 |
| 6 | Qwen3.5-4B-Base | 72.14 | #798 |
| 7 | Llama 3.1 8B | 72.09 | #1151 |
| 8 | Qwen 3 14B (Non-reasoning) | 71.72 | #537 (Qwen 3 14B) |
| 9 | gemma-3-12B-pt | 71.29 | #686 |
| 10 | Qwen 2.5 72B Instruct | 70.95 | #454 |
| 11 | Qwen 3 32B (Non-reasoning) | 70.16 | #443 (Qwen 3 32B) |
| 12 | Llama 3.1 8B Instruct | 69.09 | #1035 |
| 13 | gemma-3-4B-pt Base | 68.69 | #1134 |
| 14 | Qwen 3 14B | 68.35 | #537 |
| 15 | Ministral-3-3B-Reasoning-2512 | 68.2 | #932 |
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-multiwikiqa-lt · How It Works · Data refreshed daily, snapshot 2026-10-20.