EuroEval Latvian Knowledge: leaderboard
Metric: Knowledge Average Score (%). Source: euroeval.com. 212 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 (Thinking) | 87.47 |
| 2 | Gemini 2.5 Pro | 87.46 |
| 3 | GPT-5 | 86.7 |
| 4 | Gemini 3 Pro (Preview) | 84.63 |
| 5 | GPT-5.4 Mini (Medium) | 84.53 |
| 6 | Grok 4.1 Fast (Reasoning) | 84.18 |
| 7 | GPT-5.4 Mini (High) | 83.71 |
| 8 | Gemini 2.5 Flash (Thinking) | 83.57 |
| 9 | GPT-5.2 | 82.23 |
| 10 | GPT-5 Mini | 82.15 |
| 11 | Gemini 3.1 Flash Lite (Preview) | 79.95 |
| 12 | Gemma 4 31B (IT) | 79.32 |
| 13 | GPT-5.4 Nano (Medium) | 77.73 |
| 14 | Claude Sonnet 4.6 | 77.02 |
| 15 | Qwen 3.5 9B | 77.02 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-latvian-knowledge · How It Works · Data refreshed daily, snapshot 2026-09-05.