Evals for Every Language - Welsh (cy): leaderboard
Average score across evaluation tasks in Welsh (cy), from Evals for Every Language.
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 66 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 84.91 | #58 |
| 2 | Gemini 3.1 Flash Lite | 84.88 | #191 |
| 3 | Claude Sonnet 4.5 | 84.75 | #142 |
| 4 | Claude Sonnet 4 | 84.41 | #199 |
| 5 | Claude Opus 4.7 | 83.6 | #49 |
| 6 | Claude Opus 4.8 | 83.6 | #37 |
| 7 | Grok 4.20 | 81.72 | #175 |
| 8 | GPT-5 | 81.66 | #98 |
| 9 | GPT-5.1 | 81.17 | #135 |
| 10 | GPT-5.3 | 80.63 | #103 |
| 11 | Claude Haiku 4.5 | 80.57 | #280 |
| 12 | GPT-5.5 | 80.43 | #31 |
| 13 | GPT-5.2 | 80.34 | #110 |
| 14 | Claude Sonnet 4.6 | 80.2 | #90 |
| 15 | GPT-4o | 80.03 | #343 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-cy · How It Works · Data refreshed daily, snapshot 2026-10-20.