granite-8B-code-instruct-4k: benchmark results
Provider: IBM. Released 2024-05-06. Access: Open.
Aggregate Rating 1283 ± 37, rank #8947 of 10916 rated models, from 34 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Portuguese NLU - HAREM | 47.56 | Micro F1 excluding MISC (%) | 71.4 |
| EuroEval Italian NLU - MultiNERD IT | 65.53 | Micro F1 excluding MISC (%) | 61.1 |
| EuroEval Spanish NLU - CoNLL ES | 61.24 | Micro F1 excluding MISC (%) | 52.6 |
| EuroEval Portuguese NLU | 48.91 | NLU Average Score (%) | 47.3 |
| EuroEval Spanish NLU | 40.87 | NLU Average Score (%) | 45.6 |
| EuroEval Spanish NLU - Sentiment Headlines ES | 37.42 | Matthews correlation coefficient (%) | 45.6 |
| EuroEval Spanish NLU - MLQA ES | 58.44 | Character F1 (%) | 45.2 |
| EuroEval Portuguese NLU - SST-2 PT | 74.37 | Matthews correlation coefficient (%) | 44.3 |
| EuroEval Portuguese NLU - ScaLA PT | 8.35 | Matthews correlation coefficient (%) | 42 |
| EuroEval Spanish NLU - ScaLA ES | 6.37 | Matthews correlation coefficient (%) | 41.4 |
| EuroEval Portuguese NLU - MultiWikiQA PT | 65.35 | Character F1 (%) | 38.9 |
| EuroEval Italian NLU - SQuAD IT | 59.23 | Character F1 (%) | 38.6 |
Interactive version: theaggregate.ai/model?slug=granite-8b-code-instruct-4k · How It Works · Data refreshed daily, snapshot 2026-11-08.