Evals for Every Language - Chuvash (cv): leaderboard
Average score across evaluation tasks in Chuvash (cv), from Evals for Every Language.
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Around January 2027. 64 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 70.19 | #58 |
| 2 | Claude Opus 4.8 | 70.16 | #37 |
| 3 | Claude Opus 4.7 | 68.23 | #49 |
| 4 | GPT-5.3 | 67.18 | #103 |
| 5 | Qwen 3.6 Plus | 67.15 | #120 |
| 6 | Gemini 2.5 Pro | 65.34 | #152 |
| 7 | Grok 4.20 | 65.34 | #175 |
| 8 | Claude Sonnet 4.6 | 64.38 | #90 |
| 9 | GPT-5.4 | 64.01 | #79 |
| 10 | GPT-5.5 | 63.84 | #31 |
| 11 | Gemini 3.1 Flash Lite | 61.66 | #191 |
| 12 | DeepSeek V3.1 | 60.57 | #264 |
| 13 | Claude Sonnet 4.5 | 60.44 | #142 |
| 14 | GPT-5.2 | 60.41 | #110 |
| 15 | Nova Pro (v1) | 59.55 | #497 |
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-cv · How It Works · Data refreshed daily, snapshot 2026-10-20.