Evals for Every Language - Language pa: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 77.29 |
| 2 | Claude Opus 4.8 | 76.01 |
| 3 | GPT-5.5 | 75.93 |
| 4 | Qwen 3.6 Plus | 75.68 |
| 5 | GPT-5 | 75.59 |
| 6 | GPT-5.1 | 75.12 |
| 7 | Gemini 3.1 Flash Lite | 74.71 |
| 8 | Claude Opus 4.7 | 74.43 |
| 9 | GPT-5.3 | 74.37 |
| 10 | GPT-5.2 | 73.96 |
| 11 | Grok 4.20 | 73.79 |
| 12 | Claude Sonnet 4.6 | 73.37 |
| 13 | Nova Pro (v1) | 73.23 |
| 14 | Gemini 2.5 Flash Lite | 72.87 |
| 15 | Mistral Medium 3.1 | 72.81 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-pa · How It Works · Data refreshed daily, snapshot 2026-09-05.