Open Persian LLM - Part Multiple Choice — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 31 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.3 70B Instruct | 51.42 |
| 2 | Gemma 3 27B (IT) | 48.56 |
| 3 | QwQ 32B-Preview | 46.64 |
| 4 | Qwen 2.5 32B Instruct | 46.06 |
| 5 | Gemma 2 27B (IT) | 46.03 |
| 6 | Gemma 3 12B (IT) | 43.52 |
| 7 | aya-expanse-32B | 43.36 |
| 8 | Gemma 2 9B (IT) | 42.7 |
| 9 | aya-23-35B | 36.79 |
| 10 | Qwen 2.5 7B Instruct | 36.72 |
| 11 | Llama 3.1 8B Instruct | 36.68 |
| 12 | Qwen 2 7B Instruct | 35.9 |
| 13 | Llama 3.1 8B | 35.52 |
| 14 | Hermes-3-Llama-3.1-8B | 35.01 |
| 15 | Llama 3 8B Instruct | 34.99 |
Interactive version: theaggregate.ai/benchmark?slug=open-persian-llm-part-multiple-choice · How the rankings work · Data refreshed daily, snapshot 2026-07-22.