Evals for Every Language - Language sat — leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 44.28 |
| 2 | Gemini 2.5 Flash | 43.82 |
| 3 | Gemini 2.5 Pro | 43.58 |
| 4 | Claude Sonnet 4.5 | 43.24 |
| 5 | Claude Sonnet 4 | 42.87 |
| 6 | Gemini 3.1 Flash Lite | 41.4 |
| 7 | Qwen 3.6 Plus | 40.82 |
| 8 | Claude Opus 4.7 | 40.21 |
| 9 | Claude Opus 4.8 | 39.68 |
| 10 | DeepSeek V3.2 Exp | 37.8 |
| 11 | Nova Pro (v1) | 37.12 |
| 12 | Gemini 3.1 Pro (Preview) | 36.93 |
| 13 | GPT-5.4 | 36.57 |
| 14 | Llama 4 Maverick | 35.24 |
| 15 | DeepSeek V3.1 | 34.7 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-sat · How the rankings work · Data refreshed daily, snapshot 2026-07-22.