Granite 3.3 8B (Non-reasoning): benchmark results
Provider: IBM. Released 2025-02-26. Access: Open.
Unified ELO 1390 ± 1, rank #1408 of 1761 rated models, from 17 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA CritPt | 0 | Accuracy (%) | 24.4 |
| AA Omniscience - Law | 7.3 | Accuracy (%) | 18.3 |
| AA Humanity's Last Exam | 4.17 | Accuracy (%) | 16.7 |
| CritPt | 0 | Accuracy (self-reported) | 16 |
| AA Omniscience - Health | 10.3 | Accuracy (%) | 12.2 |
| AA Omniscience - Software Engineering (SWE) | 12.4 | Accuracy (%) | 10.8 |
| AA GPQA Diamond | 33.84 | Accuracy (%) | 10.5 |
| AA Omniscience - Humanities & Social Sciences | 9.6 | Accuracy (%) | 10.2 |
| AA-Omniscience Accuracy | 9.65 | Accuracy (%) | 10.2 |
| AA Omniscience - Business | 8 | Accuracy (%) | 9.8 |
| Artificial Analysis Intelligence Index | 1 | Intelligence Index | 9.4 |
| BenchmarkList ECI | 70.55 | Capability Index (ECI) | 7.8 |
Interactive version: theaggregate.ai/model?slug=granite-3-3-8b-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.