Bielik-11B-v3-Base-20250730: benchmark results
Provider: SpeakLeash. Released 2025-07-30. Access: Open.
Aggregate Rating 1509 ± 25, rank #2222 of 10916 rated models, from 269 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open PL LLM - PolQA Closed Book (generative, 5-shot) | 87.12 | Levenshtein Similarity (%) | 100 |
| Open PL LLM - 8TAGS (generative, 5-shot) | 80.01 | Accuracy (%) | 98.7 |
| EuroEval Albanian NLU - MultiWikiQA SQ | 69.27 | Character F1 (%) | 98.5 |
| EuroEval Slovak NLU - Multi Wiki QA SK | 74.17 | Character F1 (%) | 98 |
| EuroEval Lithuanian NLU - MultiWikiQA LT | 73.7 | Character F1 (%) | 97.8 |
| EuroEval Romanian NLU - MultiWikiQA RO | 76.33 | Character F1 (%) | 97.6 |
| EuroEval Hungarian NLU - MultiWikiQA HU | 69.98 | Character F1 (%) | 97.3 |
| EuroEval Latvian NLU - MultiWikiQA LV | 70.61 | Character F1 (%) | 96.6 |
| EuroEval Polish NLU - PoQuAD | 63.05 | Character F1 (%) | 96.6 |
| EuroEval Catalan NLU - MultiWikiQA CA | 76.85 | Character F1 (%) | 96.2 |
| Open PL LLM - PolQA Open Book (generative, 5-shot) | 91.75 | Levenshtein Similarity (%) | 96.2 |
| EuroEval Greek NLU - Greek SA | 81.91 | Matthews correlation coefficient (%) | 96.1 |
Interactive version: theaggregate.ai/model?slug=bielik-11b-v3-base-20250730 · How It Works · Data refreshed daily, snapshot 2026-11-08.