Evals for Every Language - Odia (or): leaderboard
Average score across evaluation tasks in Odia (or), 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 | 76.13 | #49 |
| 2 | Claude Sonnet 4.5 | 75.22 | #142 |
| 3 | Claude Sonnet 4 | 75.17 | #199 |
| 4 | Gemini 3.1 Pro (Preview) | 74.36 | #58 |
| 5 | Claude Opus 4.8 | 73.6 | #37 |
| 6 | GPT-5.1 | 73.24 | #135 |
| 7 | GPT-4o | 72.01 | #343 |
| 8 | DeepSeek V3.1 | 71.88 | #264 |
| 9 | Qwen 3.6 Plus | 71.39 | #120 |
| 10 | Grok 4.20 | 71.27 | #175 |
| 11 | GPT-5.5 | 70.82 | #31 |
| 12 | Claude Sonnet 4.6 | 70.51 | #90 |
| 13 | GPT-5 | 70.35 | #98 |
| 14 | GPT-5.4 | 68.94 | #79 |
| 15 | GPT-5.2 | 68.46 | #110 |
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-or · How It Works · Data refreshed daily, snapshot 2026-10-20.