EuroEval Latvian NLU - MultiWikiQA LV — leaderboard
Metric: Reading comprehension Score (%). Source: euroeval.com. 233 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | gemma-3-27B-pt | 73.01 |
| 2 | Mistral Small 3.2 | 72.5 |
| 3 | Qwen 3 14B (Non-reasoning) | 72.18 |
| 4 | Llama 3.1 70B | 72.11 |
| 5 | Magistral Small | 71.87 |
| 6 | Ministral-3-14B-Reasoning-2512 | 70.52 |
| 7 | Mistral Small 3.1 | 69.2 |
| 8 | Ministral 3 14B | 68.56 |
| 9 | Ministral-3-3B-Reasoning-2512 | 68.05 |
| 10 | Qwen 2.5 72B Instruct | 68.04 |
| 11 | Qwen 3 14B | 67.91 |
| 12 | gemma-3-12B-pt | 67.15 |
| 13 | Gemma 4 31B (IT) | 66.57 |
| 14 | Ministral 3 8B | 66.3 |
| 15 | Llama 3.1 8B | 66.24 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-latvian-nlu-multiwikiqa-lv · How the rankings work · Data refreshed daily, snapshot 2026-07-22.