Evals for Every Language - Language sc: leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Flash Lite | 60 |
| 2 | Claude Opus 4.7 | 58.44 |
| 3 | Gemini 3.1 Pro (Preview) | 57.52 |
| 4 | Claude Sonnet 4.5 | 56.09 |
| 5 | Gemini 2.5 Flash | 55.96 |
| 6 | Gemini 2.5 Pro | 55.38 |
| 7 | Claude Sonnet 4 | 54.87 |
| 8 | Gemini 2.5 Flash Lite | 54.55 |
| 9 | GPT-5.4 | 54.08 |
| 10 | Claude Opus 4.8 | 52.53 |
| 11 | GPT-5.3 | 51.49 |
| 12 | GPT-5 | 51.21 |
| 13 | Grok 4.20 | 50.5 |
| 14 | GPT-5.5 | 50.43 |
| 15 | Llama 4 Maverick | 50.02 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-sc · How It Works · Data refreshed daily, snapshot 2026-09-05.