granite-3.2-8B-instruct-preview: benchmark results
Provider: IBM. Released 2025-02-07. Access: Open.
Aggregate Rating 1436 ± 34, rank #4360 of 10916 rated models, from 56 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Japanese LLM - Xlsum JA Rouge1 | 35.61 | Score (%) | 94.5 |
| Open Japanese LLM - Xlsum JA RougeLsum | 28.98 | Score (%) | 93.3 |
| Open Japanese LLM - Xlsum JA Bert Score JA F1 | 71.13 | Score (%) | 92.5 |
| Enkrypt AI - Toxicity Risk | 0.41 | Risk Score | 91.5 |
| Open Japanese LLM - SUM | 11.88 | Score (%) | 91.1 |
| Open Japanese LLM - Xlsum JA Rouge2 | 11.88 | Score (%) | 91.1 |
| Open Japanese LLM - Xlsum JA Bleu JA | 3.07 | Score (%) | 83.1 |
| Enkrypt AI - Safety Risk | 23.6 | Risk Score | 66.2 |
| Open Japanese LLM - ALT J TO E Bleu EN | 16.91 | Score (%) | 62.9 |
| Open Japanese LLM - Wiki NER SET F1 | 6.19 | Score (%) | 57.2 |
| Enkrypt AI - Jailbreak Risk | 8.16 | Risk Score | 57 |
| Open Japanese LLM - Wiki Coreference SET F1 | 3.35 | Score (%) | 55.8 |
Interactive version: theaggregate.ai/model?slug=granite-3-2-8b-instruct-preview · How It Works · Data refreshed daily, snapshot 2026-11-08.