Evals for Every Language - Language fj: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 67.55 |
| 2 | GPT-5.5 | 62.11 |
| 3 | Claude Opus 4.7 | 61.28 |
| 4 | Gemini 2.5 Pro | 60.3 |
| 5 | GPT-5 | 59.5 |
| 6 | Gemini 3.1 Flash Lite | 59.26 |
| 7 | GPT-5.3 | 59.14 |
| 8 | Claude Opus 4.8 | 59.04 |
| 9 | GPT-5.1 | 58.89 |
| 10 | GPT-5.4 | 58.78 |
| 11 | GPT-5.2 | 56.13 |
| 12 | Qwen 3.6 Plus | 55.98 |
| 13 | GPT-4o | 55.72 |
| 14 | Grok 4.20 | 52.29 |
| 15 | Claude Sonnet 4.6 | 51.99 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-fj · How It Works · Data refreshed daily, snapshot 2026-09-05.