Qwen 1.5 0.5B — benchmark results
Provider: Alibaba. Released 2024-02-04. Access: Open.
Unified ELO 1149 ± 27, rank #1736 of 1776 rated models, from 112 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Japanese LLM - Wikicorpus J TO E Bleu EN | 21.55 | Score (%) | 96.9 |
| Open Japanese LLM - Mbpp Pylint Check | 14.26 | Score (%) | 42.7 |
| Open Japanese LLM - CG | 1.41 | Score (%) | 36.8 |
| EuroEval Spanish NLU - ScaLA ES | 1.58 | Linguistic acceptability Score (%) | 30.4 |
| SALAD-Bench Attack | 7.78 | Safety Score (%) | 30.3 |
| EuroEval Faroese NLU - ScaLA FO | 0.45 | Linguistic acceptability Score (%) | 24.7 |
| Open LLM Leaderboard - MuSR | 4.3 | Score | 21.2 |
| SALAD-Bench | 44.07 | Average Safety Score (%) | 21.2 |
| SALAD-Bench Base | 80.36 | Safety Score (%) | 21.2 |
| EuroEval Portuguese NLU - ScaLA PT | 1 | Linguistic acceptability Score (%) | 19 |
| EuroEval Norwegian Common Sense Reasoning | 1.59 | Common Sense Reasoning Average Score (%) | 18 |
| Open Japanese LLM - Xlsum JA Rouge1 | 17.52 | Score (%) | 18 |
Interactive version: theaggregate.ai/model?slug=qwen-1-5-0-5b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.