Granite 4.1 8B: benchmark results
IBM's Apache-2.0 8B dense Granite 4.1 instruct model (April 2026) with 512K context, matching the prior 32B MoE Granite 4.0-H-Small. Provider: IBM. Released 2026-04-29. Access: Open.
Aggregate Rating 1461 ± 20, rank #930 of 1656 rated models, from 121 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SpeechMap Compliance | 84 | % Requests Completed | 82.2 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 18 | Accuracy (%) | 55.3 |
| AA Omniscience - Software Engineering (SWE) - PHP | 22 | Accuracy (%) | 55.3 |
| BLXBench | 64.6 | Score (imported) | 54.2 |
| AA Omniscience - Software Engineering (SWE) - Swift | 32 | Accuracy (%) | 51.2 |
| Chatbot Arena (Text - Chinese) | 1367 | Arena Score | 43.1 |
| AA Omniscience - Software Engineering (SWE) - Rust | 44 | Accuracy (%) | 43 |
| Chatbot Arena (Text - Math) | 1317 | Arena Score | 42.2 |
| Chatbot Arena (Text - Mathematical) | 1325 | Arena Score | 41.4 |
| Chatbot Arena (Text - Legal & Government) | 1345 | Arena Score | 40.2 |
| Chatbot Arena (Text - Software & IT Services) | 1352 | Arena Score | 38.9 |
| Chatbot Arena (Text - Expert) | 1338 | Arena Score | 37.3 |
Interactive version: theaggregate.ai/model?slug=granite-4-1-8b · How It Works · Data refreshed daily, snapshot 2026-10-20.