babbage (1.3B): benchmark results
Provider: Other.
Aggregate Rating 1049 ± 38, rank #10910 of 10916 rated models, from 32 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM Classic - LSAT | 26.09 | Exact Match (%) | 100 |
| HELM Classic - BLiMP | 83.28 | Exact Match (%) | 83.9 |
| RealToxicityPrompts | 0.05 | Toxic fraction | 51.2 |
| HELM Classic - Dyck | 47.27 | Exact Match (%) | 33.8 |
| HELM Classic - Synthetic Reasoning Natural | 11.99 | F1 (%) | 33.8 |
| HELM Classic - Entity Matching | 63.16 | Exact Match (%) | 28.8 |
| HELM Classic - CivilComments | 51.9 | Exact Match (%) | 27.3 |
| HELM Classic - Synthetic Reasoning Abstract | 15.99 | Exact Match (%) | 26.5 |
| HELM Classic - LegalSupport | 49.22 | Exact Match (%) | 25.7 |
| HELM Classic - WikiFact | 18.38 | Exact Match (%) | 19.7 |
| HELM Classic - MATH | 4.83 | Equivalent (%) | 16.9 |
| HELM Classic - QuAC | 27.26 | F1 (%) | 16.9 |
Interactive version: theaggregate.ai/model?slug=babbage-1-3b · How It Works · Data refreshed daily, snapshot 2026-11-08.