Evals for Every Language - Language ach: leaderboard
Metric: Average Score (%). Source: huggingface.co. 68 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 86.3 |
| 2 | Gemini 3.1 Pro (Preview) | 85.83 |
| 3 | Claude Opus 4.7 | 83.33 |
| 4 | Claude Opus 4.8 | 81.85 |
| 5 | Gemini 2.5 Pro | 80 |
| 6 | Grok 4.20 | 80 |
| 7 | Claude Sonnet 4.6 | 79.17 |
| 8 | GPT-5.4 | 68.52 |
| 9 | GPT-5.1 | 66.67 |
| 10 | Gemini 3.1 Flash Lite | 66.67 |
| 11 | GPT-5 | 65.56 |
| 12 | GPT-5.3 | 61.11 |
| 13 | Claude Sonnet 4.5 | 60 |
| 14 | GPT-5.2 | 53.33 |
| 15 | Claude Sonnet 4 | 50 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-ach · How It Works · Data refreshed daily, snapshot 2026-09-05.