Cohere large v20220720 (13.1B): benchmark results
Provider: Other.
Aggregate Rating 1156 ± 34, rank #9865 of 10916 rated models, from 32 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM Classic - BLiMP | 83.35 | Exact Match (%) | 87.1 |
| RealToxicityPrompts | 0.01 | Toxic fraction | 80.5 |
| HELM Classic - WikiFact | 28.55 | Exact Match (%) | 60.6 |
| HELM Classic - OpenbookQA | 54.2 | Exact Match (%) | 59.7 |
| HELM Classic - MMLU | 32.36 | Exact Match (%) | 47 |
| HELM Classic - XSUM | 10.82 | ROUGE-2 (%) | 46.3 |
| HELM Classic - Dyck | 53.13 | Exact Match (%) | 45.6 |
| HELM Classic - HellaSwag | 73.6 | Exact Match (%) | 45.2 |
| HELM Classic - IMDB | 93.33 | Exact Match (%) | 44.7 |
| HELM Classic - QuAC | 33.84 | F1 (%) | 44.6 |
| HELM Classic - BoolQ | 72.53 | Exact Match (%) | 43.9 |
| HELM Classic - RAFT | 59.62 | Exact Match (%) | 40.9 |
Interactive version: theaggregate.ai/model?slug=cohere-large-v20220720-13-1b · How It Works · Data refreshed daily, snapshot 2026-11-08.