Evals for Every Language - Language tk: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 70.31 |
| 2 | Claude Opus 4.7 | 70.15 |
| 3 | Claude Opus 4.8 | 69.26 |
| 4 | Claude Sonnet 4.6 | 68.44 |
| 5 | GPT-5.3 | 68.32 |
| 6 | Claude Sonnet 4.5 | 67.46 |
| 7 | GPT-5.5 | 65.91 |
| 8 | GPT-5.2 | 65.42 |
| 9 | Grok 4.20 | 65.24 |
| 10 | GPT-5 | 64.69 |
| 11 | Gemini 2.5 Pro | 64.63 |
| 12 | GPT-4o | 64.01 |
| 13 | Claude Sonnet 4 | 63.9 |
| 14 | GPT-5.1 | 63.67 |
| 15 | Qwen 3.6 Plus | 63.17 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-tk · How It Works · Data refreshed daily, snapshot 2026-09-05.