EuroEval Finnish: leaderboard
Metric: Average Score (%). Source: euroeval.com. 382 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 (High) | 72.92 |
| 2 | GPT-5 | 71.55 |
| 3 | Gemini 2.5 Pro | 69.81 |
| 4 | GPT-5 (Minimal) | 69.28 |
| 5 | Gemini 2.5 Flash (Thinking) | 68.93 |
| 6 | Gemini 2.5 Flash (Preview 04-17) | 68.77 |
| 7 | Qwen 3 235B A22B 2507 Instruct | 67.94 |
| 8 | O3 (2025-04-16) | 67.87 |
| 9 | Llama 3.1 405B | 67.48 |
| 10 | Gemma 4 31B (IT) | 66.55 |
| 11 | GPT-5 Mini (High) | 66.43 |
| 12 | Gemini 3 Pro (Preview) | 65.88 |
| 13 | Claude 3.7 Sonnet (20250219) (Thinking) | 65.82 |
| 14 | GPT-4o (2024-05-13) | 65.79 |
| 15 | GPT-5 Nano (High) | 65.28 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-finnish · How It Works · Data refreshed daily, snapshot 2026-09-05.