Open Persian LLM - ARC Challenge — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 31 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 3 27B (IT) | 90.6 |
| 2 | Llama 3.3 70B Instruct | 89.93 |
| 3 | QwQ 32B-Preview | 87.24 |
| 4 | Gemma 2 27B (IT) | 85.91 |
| 5 | Qwen 2.5 32B Instruct | 85.91 |
| 6 | Gemma 2 9B (IT) | 84.56 |
| 7 | Gemma 3 12B (IT) | 81.21 |
| 8 | aya-expanse-32B | 79.87 |
| 9 | aya-23-35B | 72.48 |
| 10 | aya-expanse-8B | 70.47 |
| 11 | Qwen 2.5 7B Instruct | 69.13 |
| 12 | Qwen 2 7B Instruct | 68.46 |
| 13 | aya-23-8B | 64.43 |
| 14 | Gemma 3 4B (IT) | 63.76 |
| 15 | Llama 3.1 8B Instruct | 60.4 |
Interactive version: theaggregate.ai/benchmark?slug=open-persian-llm-arc-challenge · How the rankings work · Data refreshed daily, snapshot 2026-07-22.