granite-7B-base: benchmark results
Provider: IBM. Released 2024-04-19. Access: Open.
Aggregate Rating 1249 ± 34, rank #9261 of 10916 rated models, from 23 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Portuguese NLU - MultiWikiQA PT | 67.56 | Character F1 (%) | 46.1 |
| EuroEval Portuguese NLU - SST-2 PT | 73.18 | Matthews correlation coefficient (%) | 42 |
| EuroEval Dutch NLU - DBRD | 82.39 | Matthews correlation coefficient (%) | 36.7 |
| EuroEval Portuguese NLU | 42.69 | NLU Average Score (%) | 33.9 |
| EuroEval Norwegian Knowledge | 4.69 | Knowledge Average Score (%) | 28 |
| EuroEval Portuguese | 30.41 | Average Score (%) | 27.6 |
| EuroEval English Knowledge | 42.39 | Knowledge Average Score (%) | 27.4 |
| EuroEval Norwegian Knowledge - NRK Quiz QA | 9.06 | MCC (x100) | 26.8 |
| EuroEval Norwegian Knowledge - Idioms NO | 0.31 | MCC (x100) | 24.2 |
| EuroEval Danish Knowledge | 11.68 | Knowledge Average Score (%) | 22.7 |
| EuroEval Portuguese NLU - HAREM | 29.7 | Micro F1 excluding MISC (%) | 22.1 |
| EuroEval Portuguese Knowledge | 12.04 | Knowledge Average Score (%) | 21.9 |
Interactive version: theaggregate.ai/model?slug=granite-7b-base · How It Works · Data refreshed daily, snapshot 2026-11-08.