EuroEval Finnish NLU - Tydiqa FI: leaderboard
The "Tydiqa FI" subtask of EuroEval Finnish NLU. Finnish-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: Not forecast. 374 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen3.5-4B-Base | 77.51 | #798 |
| 2 | EuroLLM-9B | 77.33 | #1254 |
| 3 | Gemma 2 27B | 77.11 | #806 |
| 4 | Llama 3 70B | 77.11 | #631 |
| 5 | Llama 3.1 70B | 76.96 | #592 |
| 6 | Ministral-3-14B-Reasoning-2512 | 76.61 | #540 |
| 7 | Magistral Small 1.2 | 76.47 | #570 |
| 8 | Llama 2 70B Base | 76.05 | #1025 |
| 9 | Yi 1.5 34B | 75.98 | #807 |
| 10 | SOLAR-10.7B-v1.0 | 75.92 | #1253 |
| 11 | Gemma 3n E4B | 75.84 | #878 |
| 12 | gemma-3-27B-pt | 75.66 | #502 |
| 13 | Mistral Small 3.2 | 74.97 | #605 |
| 14 | Bielik-11B-v2.3-Instruct | 74.94 | #693 |
| 15 | Mistral Small 3.1 | 74.86 | #610 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-finnish-nlu-tydiqa-fi · How It Works · Data refreshed daily, snapshot 2026-10-20.