Evals for Every Language - Armenian (hy): leaderboard
Average score across evaluation tasks in Armenian (hy), from Evals for Every Language.
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Around January 2027. 66 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 75.66 | #58 |
| 2 | Claude Sonnet 4.5 | 75.11 | #142 |
| 3 | Claude Sonnet 4.6 | 74.95 | #90 |
| 4 | GPT-5 | 74.51 | #98 |
| 5 | Claude Sonnet 4 | 74.14 | #199 |
| 6 | Nova Pro (v1) | 74.14 | #497 |
| 7 | Claude Opus 4.7 | 74 | #49 |
| 8 | GPT-5.3 | 73.45 | #103 |
| 9 | Claude Opus 4.8 | 72.9 | #37 |
| 10 | Qwen 3.6 Plus | 72.81 | #120 |
| 11 | GPT-5.5 | 72.29 | #31 |
| 12 | Qwen 3.6 Flash | 72.24 | #214 |
| 13 | Grok 4.20 | 72.19 | #175 |
| 14 | GPT-5.2 | 72.09 | #110 |
| 15 | DeepSeek V3.1 | 71.94 | #264 |
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-hy · How It Works · Data refreshed daily, snapshot 2026-10-20.