Evals for Every Language - Language ast — leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 64.58 |
| 2 | Claude Sonnet 4.5 | 64.11 |
| 3 | Gemini 2.5 Flash | 63.8 |
| 4 | Claude Opus 4.8 | 63.78 |
| 5 | Claude Opus 4.7 | 63.21 |
| 6 | Gemini 2.5 Pro | 63.1 |
| 7 | Claude Sonnet 4 | 62.24 |
| 8 | Nova Pro (v1) | 60.61 |
| 9 | Claude Sonnet 4.6 | 60.53 |
| 10 | GPT-5.4 | 60.43 |
| 11 | Gemma 3 27B (IT) | 59.58 |
| 12 | DeepSeek V3.2 Exp | 59.41 |
| 13 | GPT-4o | 59.35 |
| 14 | Llama 4 Maverick | 58.82 |
| 15 | GPT-5.2 | 58.29 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-ast · How the rankings work · Data refreshed daily, snapshot 2026-07-22.