Bielik-11B-v2.6-Instruct: benchmark results
Provider: SpeakLeash. Released 2025-04-01. Access: Open.
Aggregate Rating 1529 ± 17, rank #1963 of 10916 rated models, from 336 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open PL LLM - PSC (generative, 5-shot) | 98.54 | Binary F1 (%) | 97 |
| Open PL LLM - PolQA Closed Book (generative, 5-shot) | 83.8 | Levenshtein Similarity (%) | 95.8 |
| Open PL LLM - PolEmo2-OUT (generative, 5-shot) | 79.96 | Accuracy (%) | 95.4 |
| Polish EQ-Bench | 73.7 | EQ-Bench Score (parseability-adjusted) | 95 |
| Open PL LLM - EQ-Bench | 61.89 | EQ-Bench Score | 94.8 |
| Open PL LLM - PolQA Open Book (generative, 5-shot) | 91.16 | Levenshtein Similarity (%) | 94.8 |
| Open PL LLM - PolEmo2-OUT (multiple choice, 5-shot) | 79.55 | Accuracy (%) | 94.2 |
| EuroEval Polish NLU - ScaLA PL | 57.38 | Matthews correlation coefficient (%) | 93.1 |
| EuroEval Ukrainian NLU - Cross Domain UK Reviews | 62.54 | Matthews correlation coefficient (%) | 93 |
| EuroEval Spanish NLU - ScaLA ES | 43.62 | Matthews correlation coefficient (%) | 92.7 |
| EuroEval Croatian NLU - MMS HR | 44.48 | Matthews correlation coefficient (%) | 91.9 |
| EuroEval Polish NLU - Polemo2 | 93.93 | Matthews correlation coefficient (%) | 91.4 |
Interactive version: theaggregate.ai/model?slug=bielik-11b-v2-6-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.