Evals for Every Language - Language ts: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 71.05 |
| 2 | GPT-5.5 | 68.56 |
| 3 | Gemini 2.5 Pro | 66.48 |
| 4 | GPT-5.4 | 64.96 |
| 5 | GPT-5 | 64.94 |
| 6 | Gemini 3.1 Flash Lite | 64.58 |
| 7 | GPT-5.1 | 64.38 |
| 8 | GPT-5.3 | 63.67 |
| 9 | Claude Opus 4.8 | 63.45 |
| 10 | Claude Opus 4.7 | 61.92 |
| 11 | Claude Sonnet 4.5 | 57.7 |
| 12 | Grok 4.20 | 56.65 |
| 13 | GPT-5.2 | 56.64 |
| 14 | GPT-4o | 56.14 |
| 15 | Qwen 3.6 Plus | 55.93 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-ts · How It Works · Data refreshed daily, snapshot 2026-09-05.