Open PL LLM Leaderboard: leaderboard
Polish language LLM evaluation across Polemo2, 8tags, Belebele, Polish QA, EQ-Bench, RAG, and other Polish understanding and generation tasks.
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Around November 2026. 287 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Mistral Large 2 (Nov) Instruct (2411) | 69.84 | #465 |
| 2 | Llama 3.1 405B Instruct FP8 | 69.44 | #389 |
| 3 | Mistral Large 2 (Jul) | 69.11 | #484 |
| 4 | Qwen 2.5 72B Instruct | 67.92 | #454 |
| 5 | Qwen 2.5 72B | 67.38 | #499 |
| 6 | QwQ 32B-Preview | 67.01 | #651 |
| 7 | Qwen 2.5 32B | 66.73 | #546 |
| 8 | Gemma 3 27B (IT) | 66.48 | #523 |
| 9 | Llama 3.3 70B Instruct | 66.4 | #534 |
| 10 | Qwen 2 72B | 66.02 | #521 |
| 11 | Qwen 2 72B Instruct | 65.87 | #576 |
| 12 | Bielik-11B-v2.3-Instruct | 65.71 | #693 |
| 13 | Bielik-11B-v2.2-Instruct | 65.57 | #660 |
| 14 | Llama 3.1 70B Instruct | 65.49 | #561 |
| 15 | Bielik-11B-v2.1-Instruct | 65.45 | #673 |
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-leaderboard · How It Works · Data refreshed daily, snapshot 2026-10-20.