Open Persian LLM Leaderboard: leaderboard
Persian language LLM evaluation: benchmarks models on Part Multiple Choice, ARC Easy/Challenge, MMLU Pro, and AUT Multiple Choice in Persian.
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 31 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 3.3 70B Instruct | 70.36 | #534 |
| 2 | Gemma 3 27B (IT) | 67.85 | #523 |
| 3 | Gemma 2 27B (IT) | 65.46 | #645 |
| 4 | QwQ 32B-Preview | 64.78 | #651 |
| 5 | Qwen 2.5 32B Instruct | 64.46 | #498 |
| 6 | Gemma 2 9B (IT) | 62.89 | #788 |
| 7 | aya-expanse-32B | 61.94 | #692 |
| 8 | Gemma 3 12B (IT) | 60.9 | #664 |
| 9 | aya-23-35B | 56.78 | #883 |
| 10 | aya-expanse-8B | 53.68 | #931 |
| 11 | Qwen 2.5 7B Instruct | 51.9 | #862 |
| 12 | Qwen 2 7B Instruct | 51.44 | #991 |
| 13 | Llama 3.1 8B Instruct | 50.14 | #1035 |
| 14 | aya-23-8B | 49.84 | #1299 |
| 15 | Gemma 3 4B (IT) | 49.06 | #992 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-persian-llm-leaderboard · How It Works · Data refreshed daily, snapshot 2026-10-20.