Lunar-4B: benchmark results
Provider: Sakalti. Access: Open.
Aggregate Rating 1441 ± 35, rank #4163 of 10916 rated models, from 50 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Japanese LLM - Xlsum JA Bert Score JA F1 | 70.55 | Score (%) | 87.4 |
| Open Japanese LLM - Xlsum JA Rouge1 | 29.88 | Score (%) | 80.5 |
| Open Japanese LLM - Xlsum JA RougeLsum | 25.28 | Score (%) | 79.6 |
| Open Japanese LLM - SUM | 10.56 | Score (%) | 70.8 |
| Open Japanese LLM - Xlsum JA Rouge2 | 10.56 | Score (%) | 70.7 |
| Open Japanese LLM - Wiki NER SET F1 | 7.08 | Score (%) | 62.7 |
| Open Japanese LLM - CG | 29.72 | Score (%) | 61.1 |
| Open Japanese LLM - Mbpp Pylint Check | 48.8 | Score (%) | 57.5 |
| Open Japanese LLM - Jsts Pearson | 86.69 | Score (%) | 39.2 |
| Open Japanese LLM - Jsts Spearman | 83.8 | Score (%) | 39.2 |
| Open Japanese LLM - MMLU EN Exact Match | 64.95 | Score (%) | 38.9 |
| Open Japanese LLM - Xlsum JA Bleu JA | 2.38 | Score (%) | 38.6 |
Interactive version: theaggregate.ai/model?slug=lunar-4b · How It Works · Data refreshed daily, snapshot 2026-11-08.