Evals for Every Language - Language gaa — leaderboard
Metric: Average Score (%). Source: huggingface.co. 67 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 92.96 |
| 2 | Claude Opus 4.8 | 91.67 |
| 3 | Gemini 3.1 Flash Lite | 84.13 |
| 4 | Claude Opus 4.7 | 83.33 |
| 5 | GPT-5.5 | 81.9 |
| 6 | Grok 4.20 | 70 |
| 7 | Gemini 2.5 Pro | 56.67 |
| 8 | GLM 4.5 Air | 55.56 |
| 9 | Claude Sonnet 4.6 | 52.78 |
| 10 | Claude Sonnet 4.5 | 50 |
| 11 | Gemma 3 27B (IT) | 46.67 |
| 12 | GPT-5.4 | 43.33 |
| 13 | GPT-5.2 | 41.11 |
| 14 | Qwen 3.6 Plus | 40.95 |
| 15 | Claude Sonnet 4 | 40 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-gaa · How the rankings work · Data refreshed daily, snapshot 2026-07-22.