Evals for Every Language - English (en): leaderboard
Average score across evaluation tasks in English (en), from Evals for Every Language.
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 68 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Flash Lite | 89 | #191 |
| 2 | Claude Opus 4.8 | 88.59 | #37 |
| 3 | Gemini 2.5 Pro | 88.52 | #152 |
| 4 | Claude Haiku 4.5 | 87.91 | #280 |
| 5 | Gemini 3.1 Pro (Preview) | 87.9 | #58 |
| 6 | Claude Sonnet 4.5 | 87.75 | #142 |
| 7 | GPT-5.5 | 87.24 | #31 |
| 8 | Claude Opus 4.7 | 87.21 | #49 |
| 9 | Grok 4.20 | 86.83 | #175 |
| 10 | Claude Sonnet 4.6 | 86.8 | #90 |
| 11 | GPT-5 | 86.74 | #98 |
| 12 | GPT-5.1 | 86.36 | #135 |
| 13 | GPT-5.3 | 85.78 | #103 |
| 14 | GPT-4o | 85.68 | #343 |
| 15 | Nova Pro (v1) | 85.48 | #497 |
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-en · How It Works · Data refreshed daily, snapshot 2026-10-20.