IBM Granite 3.3 8B Instruct — benchmark results
Provider: Other. Released 2025-04-16. Access: Open.
Unified ELO 1485 ± 64, rank #911 of 1839 rated models, from 7 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM AIR-Bench | 76 | Refusal Rate (%) | 68.6 |
| HELM Safety HarmBench | 81.2 | LM Evaluated Safety score (%) | 59.3 |
| HELM Safety Anthropic Red Team | 99.1 | LM Evaluated Safety score (%) | 58.7 |
| HELM Safety | 90.5 | Mean score (self-reported) | 42.8 |
| HELM Safety SimpleSafetyTests | 97.5 | LM Evaluated Safety score (%) | 32.6 |
| HELM Safety BBQ | 84 | BBQ accuracy (%) | 20.9 |
| HELM Safety XSTest | 90.7 | LM Evaluated Safety score (%) | 14 |
Interactive version: theaggregate.ai/model?slug=ibm-granite-3-3-8b-instruct · How It Works · Data refreshed daily, snapshot 2026-08-05.