Evals for Every Language - Hindi (hi): leaderboard
Average score across evaluation tasks in Hindi (hi), 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 Flash Lite | 81.45 | #191 |
| 2 | Claude Opus 4.7 | 81.25 | #49 |
| 3 | Gemini 3.1 Pro (Preview) | 79.6 | #58 |
| 4 | Claude Sonnet 4.6 | 79.58 | #90 |
| 5 | GPT-5 | 79.1 | #98 |
| 6 | GPT-5.3 | 78.68 | #103 |
| 7 | Claude Sonnet 4.5 | 78.32 | #142 |
| 8 | Claude Sonnet 4 | 77.87 | #199 |
| 9 | GPT-5.2 | 77.4 | #110 |
| 10 | Gemini 2.5 Pro | 77.22 | #152 |
| 11 | Claude Opus 4.8 | 76.66 | #37 |
| 12 | DeepSeek V3.2 Exp | 76.6 | #234 |
| 13 | Grok 4.20 | 76.07 | #175 |
| 14 | Nova Pro (v1) | 75.93 | #497 |
| 15 | GPT-5.1 | 75.86 | #135 |
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-hi · How It Works · Data refreshed daily, snapshot 2026-10-20.