Loading data...
granite-3.1-3B-a800m-base: benchmark results
Provider: IBM. Released 2024-12-18. Access: Open.
Aggregate Rating 1274 ± 20, rank #8909 of 10524 rated models, from 121 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
| Lara Mini Base Report (base-model protocol) - XNLI (English) (0-shot) | 56.9 | Accuracy (%), English benchmark | 99.1 |
| Lara Mini Base Report (base-model protocol) - PAWS (English) (0-shot) | 67.3 | Accuracy (%), English benchmark | 79.6 |
| Lara Mini Base Report (base-model protocol) - PIQA (0-shot) | 78.9 | Accuracy normalized by answer length (%), English benchmark | 71.3 |
| Lara Mini Base Report (base-model protocol) - HellaSwag (10-shot) | 74.2 | Accuracy normalized by answer length (%), English benchmark | 63 |
| Lara Mini Base Report (base-model protocol) - WinoGrande (5-shot) | 69.3 | Accuracy (%), English benchmark | 63 |
| Lara Mini Base Report (base-model protocol) - COPA (0-shot) | 84.4 | Accuracy (%), English benchmark | 61.1 |
| Lara Mini Base Report (base-model protocol) - MBPP+ (3-shot) | 43.9 | pass@1; must pass MBPP's own and the EvalPlus tests (%) | 40.7 |
| Lara Mini Base Report (base-model protocol) - XCOPA, All Languages (0-shot) | 55.7 | Accuracy (%) averaged over all languages/variants of the ben | 39.8 |
| Lara Mini Base Report (base-model protocol) - PAWS-X, All Languages (0-shot) | 53.7 | Accuracy (%) averaged over all languages/variants of the ben | 38.9 |
| Lara Mini Base Report (base-model protocol) - ARC-Challenge (25-shot) | 51.1 | Accuracy normalized by answer length (%), English benchmark | 37 |
| Lara Mini Base Report (base-model protocol) - MBPP (3-shot) | 35 | pass@1 (execution) (%) | 36.1 |
| Open PL LLM - PSC (multiple choice, 5-shot) | 46.7 | Binary F1 (%) | 34.1 |
Interactive version: theaggregate.ai/model?slug=granite-3-1-3b-a800m-base · How It Works · Data refreshed daily, snapshot 2026-11-22.