Bielik-1.5B-v3.0-Instruct: benchmark results
Provider: SpeakLeash. Released 2025-04-18. Access: Open.
Aggregate Rating 1367 ± 18, rank #7246 of 10916 rated models, from 327 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Spanish Summarization - Mlsum ES | 28.14 | chrF3++ (%) | 81 |
| EuroEval Polish NLU - Polemo2 | 88.99 | Matthews correlation coefficient (%) | 73.8 |
| EuroEval English Summarization - CNN Dailymail | 42.61 | chrF3++ (%) | 73.3 |
| Open PL LLM - PolQA Open Book (generative, 5-shot) | 83.32 | Levenshtein Similarity (%) | 61.7 |
| Open PL LLM - CBD (multiple choice, 5-shot) | 26.15 | Macro F1 (%) | 60.2 |
| Open PL LLM - PPC (generative, 5-shot) | 69.8 | Accuracy (%) | 58 |
| Open PL LLM - RAG | 59.45 | Average RAG Score (%) | 56.3 |
| LLMZSZL Leaderboard | 46.81 | Score | 56.1 |
| Open PL LLM - PSC (multiple choice, 5-shot) | 67.65 | Binary F1 (%) | 55.8 |
| EuroEval Polish NLU - PoQuAD | 46.25 | Character F1 (%) | 54.9 |
| EuroEval Catalan Summarization - Dacsa CA | 33.37 | chrF3++ (%) | 51.9 |
| Open PL LLM - PolQA Closed Book (generative, 5-shot) | 53.89 | Levenshtein Similarity (%) | 51.3 |
Interactive version: theaggregate.ai/model?slug=bielik-1-5b-v3-0-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.