granite-3.0-8B-instruct: benchmark results
IBM's Apache-2.0 Granite 3.0 8B instruct model (October 2024), an enterprise workhorse tuned for RAG, summarization, extraction, and tool use. Provider: IBM. Released 2024-10-21. Access: Open.
Aggregate Rating 1407 ± 23, rank #1205 of 1656 rated models, from 114 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - GPQA | 33.22 | Score | 81.4 |
| EuroEval Spanish NLU - Sentiment Headlines ES | 48.18 | Matthews correlation coefficient (%) | 81.3 |
| EuroEval Spanish NLU - MLQA ES | 64.53 | Character F1 (%) | 80.6 |
| EuroEval Swedish NLU - Swerec | 77.96 | Matthews correlation coefficient (%) | 80.6 |
| Enkrypt AI - Safety Risk | 22.83 | Risk Score | 69.7 |
| EuroEval Portuguese NLU - MultiWikiQA PT | 72.15 | Character F1 (%) | 68.4 |
| Enkrypt AI - Jailbreak Risk | 6.71 | Risk Score | 68.2 |
| EuroEval Italian NLU - SQuAD IT | 69.69 | Character F1 (%) | 66.5 |
| Enkrypt AI - Toxicity Risk | 2.23 | Risk Score | 64.6 |
| EuroEval Danish NLU - Angry Tweets | 50.78 | Matthews correlation coefficient (%) | 63.9 |
| Open LLM Leaderboard - IFEval | 53.1 | Score | 63.1 |
| Open PL LLM - PSC (generative, 5-shot) | 92.43 | Binary F1 (%) | 62.5 |
Interactive version: theaggregate.ai/model?slug=granite-3-0-8b-instruct · How It Works · Data refreshed daily, snapshot 2026-10-20.