Open PL LLM - PolQA Open Book (generative, 5-shot): leaderboard
Answering Polish PolQA questions with relevant passages supplied. Open PL LLM Leaderboard (SpeakLeash), 5-shot, generative variant: the answer is extracted from generated text; Levenshtein similarity to the reference answer (%), higher is better. Runs not available on the Hugging Face Hub are excluded.
Metric: Levenshtein Similarity (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 318 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Bielik-11B-v2.1-Instruct | 92.39 | #679 |
| 2 | Mixtral 8x22B | 92.21 | #725 |
| 3 | Bielik-11B-v2.3-Instruct | 92.19 | #700 |
| 4 | Bielik-11B-v2.2-Instruct | 92.12 | #669 |
| 5 | Llama 3 70B | 92.09 | #632 |
| 6 | Llama 4 Scout Base | 91.92 | #708 |
| 7 | Bielik-11B-v2.0-Instruct | 91.92 | #750 |
| 8 | Llama 3.1 70B | 91.68 | #588 |
| 9 | Bielik-11B-v2 | 91.46 | #1309 |
| 10 | Mistral Large 2 (Nov) Instruct (2411) | 91.45 | #480 |
| 11 | Llama 3.1 405B Instruct FP8 | 90.99 | #400 |
| 12 | Mistral-Small-24B-Base-2501 | 90.71 | #681 |
| 13 | Mistral Large 2 (Jul) | 90.57 | #490 |
| 14 | Bielik-11B-v2.5-Instruct | 90.5 | #575 |
| 15 | Mixtral 8x7B | 89.51 | #1094 |
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-polqa-open-book-generative-5-shot · How It Works · Data refreshed daily, snapshot 2026-11-05.