Open PL LLM - Multiple Choice: leaderboard
Open PL LLM aggregate over multiple-choice Polish tasks.
Metric: Average Multiple-Choice Score (%). Source: huggingface.co. Saturation forecast: Around October 2026. 287 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Mistral Large 2 (Nov) Instruct (2411) | 65.64 | #465 |
| 2 | Qwen 2.5 72B Instruct | 65.5 | #454 |
| 3 | Llama 3.1 405B Instruct FP8 | 65.44 | #389 |
| 4 | QwQ 32B-Preview | 65.26 | #651 |
| 5 | Mistral Large 2 (Jul) | 65.08 | #484 |
| 6 | Qwen 2.5 32B | 64.7 | #546 |
| 7 | Qwen 2.5 72B | 64.65 | #499 |
| 8 | Bielik-11B-v2.3-Instruct | 63.43 | #693 |
| 9 | Bielik-11B-v2.2-Instruct | 63.41 | #660 |
| 10 | Bielik-11B-v2.0-Instruct | 63.31 | #742 |
| 11 | Qwen 2 72B | 63.06 | #521 |
| 12 | Gemma 3 27B (IT) | 63.03 | #523 |
| 13 | Bielik-11B-v2.1-Instruct | 62.88 | #673 |
| 14 | Qwen 2 72B Instruct | 61.51 | #576 |
| 15 | Llama 3.3 70B Instruct | 61.42 | #534 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-pl-llm-multiple-choice · How It Works · Data refreshed daily, snapshot 2026-10-20.