Evals for Every Language - Language apc — leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 59.69 |
| 2 | Claude Sonnet 4.5 | 55.98 |
| 3 | Gemini 2.5 Pro | 55.8 |
| 4 | GPT-4o | 55.71 |
| 5 | Claude Opus 4.7 | 55.53 |
| 6 | Gemini 2.5 Flash | 55.5 |
| 7 | Llama 4 Maverick | 54.91 |
| 8 | Gemini 2.5 Flash Lite | 54.88 |
| 9 | GPT-5 | 54.67 |
| 10 | Claude Sonnet 4 | 54.04 |
| 11 | GPT-5.5 | 50.92 |
| 12 | DeepSeek V3.1 | 50.65 |
| 13 | GPT-5.2 | 50.15 |
| 14 | Gemini 3.1 Flash Lite | 50.09 |
| 15 | Gemma 3 27B (IT) | 49.99 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-apc · How the rankings work · Data refreshed daily, snapshot 2026-07-22.