Evals for Every Language - Language ig: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 74.28 |
| 2 | Gemini 2.5 Pro | 67.26 |
| 3 | Gemini 3.1 Flash Lite | 67.1 |
| 4 | GPT-5.5 | 66.27 |
| 5 | GPT-5 | 66.21 |
| 6 | Claude Opus 4.8 | 65.86 |
| 7 | Grok 4.20 | 65.2 |
| 8 | DeepSeek V3.1 | 65 |
| 9 | Claude Sonnet 4.6 | 64.25 |
| 10 | GPT-5.3 | 63.74 |
| 11 | Claude Sonnet 4 | 63.38 |
| 12 | GPT-5.1 | 62.97 |
| 13 | Claude Sonnet 4.5 | 62.94 |
| 14 | GPT-5.4 | 62.71 |
| 15 | GPT-5.2 | 62.66 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-ig · How It Works · Data refreshed daily, snapshot 2026-09-05.