EuroEval Albanian Knowledge — leaderboard
Metric: Knowledge Average Score (%). Source: euroeval.com. 182 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 95.46 |
| 2 | GPT-5 | 94 |
| 3 | Gemini 3 Pro (Preview) | 91.45 |
| 4 | Gemini 3 Flash (Preview) (Non-reasoning) | 90.74 |
| 5 | GPT-5.4 Mini (High) | 90.23 |
| 6 | Gemini 2.5 Pro | 88.2 |
| 7 | Grok 4.20 Beta (0309) (Reasoning) | 86.06 |
| 8 | GPT-5.4 Mini (Medium) | 85.57 |
| 9 | GPT-5 Nano | 84.63 |
| 10 | GPT-5.4 Nano (High) | 83.27 |
| 11 | Claude Sonnet 4.6 | 81.24 |
| 12 | Gemini 2.5 Flash (Non-reasoning) | 80.92 |
| 13 | GPT-5 Mini | 80.27 |
| 14 | Gemini 3.1 Flash Lite (Preview) | 80.13 |
| 15 | Grok 4.1 Fast (Reasoning) | 79.36 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-albanian-knowledge · How the rankings work · Data refreshed daily, snapshot 2026-07-22.