Evals for Every Language - Nepali (ne): leaderboard
Average score across evaluation tasks in Nepali (ne), 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 | Claude Opus 4.7 | 76.05 | #49 |
| 2 | Gemini 3.1 Pro (Preview) | 75.86 | #58 |
| 3 | Gemini 2.5 Pro | 74.38 | #152 |
| 4 | Claude Sonnet 4.6 | 74.09 | #90 |
| 5 | GPT-5 | 73.21 | #98 |
| 6 | Claude Sonnet 4 | 72.64 | #199 |
| 7 | Grok 4.20 | 72.13 | #175 |
| 8 | Claude Sonnet 4.5 | 71.92 | #142 |
| 9 | GPT-5.1 | 71.77 | #135 |
| 10 | Gemini 2.5 Flash Lite | 71.3 | #422 |
| 11 | Gemma 3 27B (IT) | 70.9 | #523 |
| 12 | GPT-5 Mini | 70.84 | #186 |
| 13 | Claude Opus 4.8 | 70.37 | #37 |
| 14 | GPT-5.3 | 70.13 | #103 |
| 15 | Gemini 3.1 Flash Lite | 70.04 | #191 |
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-ne · How It Works · Data refreshed daily, snapshot 2026-10-20.