IBM Granite 4.0 Micro — benchmark results
Provider: Other. Released 2025-10-02. Access: Open.
Unified ELO 1392 ± 63, rank #1275 of 1839 rated models, from 7 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM AIR-Bench | 66.1 | Refusal Rate (%) | 48.8 |
| HELM Safety Anthropic Red Team | 98.7 | LM Evaluated Safety score (%) | 45.3 |
| HELM Safety HarmBench | 69.5 | LM Evaluated Safety score (%) | 44.2 |
| HELM Safety | 85.5 | Mean score (self-reported) | 17.8 |
| HELM Safety BBQ | 77.9 | BBQ accuracy (%) | 16.3 |
| HELM Safety SimpleSafetyTests | 94.5 | LM Evaluated Safety score (%) | 16.3 |
| HELM Safety XSTest | 86.9 | LM Evaluated Safety score (%) | 8.1 |
Interactive version: theaggregate.ai/model?slug=ibm-granite-4-0-micro · How It Works · Data refreshed daily, snapshot 2026-08-05.