Evals for Every Language - Language tt: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.8 | 70.77 |
| 2 | Claude Opus 4.7 | 70.29 |
| 3 | Gemini 3.1 Pro (Preview) | 69.73 |
| 4 | GPT-5.3 | 69.52 |
| 5 | Claude Sonnet 4.5 | 69.46 |
| 6 | GPT-4o | 69.24 |
| 7 | GPT-5 | 68.39 |
| 8 | GPT-5.2 | 68.37 |
| 9 | Gemini 2.5 Pro | 68.09 |
| 10 | Qwen 3.6 Plus | 68.06 |
| 11 | GPT-5.4 | 67.49 |
| 12 | GPT-5.5 | 67.26 |
| 13 | GPT-5.1 | 66.98 |
| 14 | Grok 4.20 | 66.32 |
| 15 | Gemini 3.1 Flash Lite | 65.89 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-tt · How It Works · Data refreshed daily, snapshot 2026-09-05.