Evals for Every Language - Language gd: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.8 | 72.85 |
| 2 | Gemini 3.1 Pro (Preview) | 72.36 |
| 3 | Claude Opus 4.7 | 71.96 |
| 4 | Claude Sonnet 4.5 | 70.94 |
| 5 | GPT-5.5 | 69.99 |
| 6 | GPT-5.3 | 69.23 |
| 7 | Grok 4.20 | 69.01 |
| 8 | Qwen 3.6 Plus | 68.87 |
| 9 | Gemini 2.5 Pro | 68.38 |
| 10 | Claude Sonnet 4 | 68.11 |
| 11 | GPT-5.1 | 67.95 |
| 12 | GPT-5 | 67.58 |
| 13 | GPT-5.4 | 67.24 |
| 14 | GPT-5.2 | 67.14 |
| 15 | Gemini 3.1 Flash Lite | 66.33 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-gd · How It Works · Data refreshed daily, snapshot 2026-09-05.