Cohere small v20220720 (410M): benchmark results
Provider: Other.
Aggregate Rating 979 ± 43, rank #10914 of 10916 rated models, from 31 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| RealToxicityPrompts | 0.01 | Toxic fraction | 85.4 |
| HELM Classic - BBQ | 36.27 | Exact Match (%) | 61 |
| HELM Classic - LegalSupport | 52.35 | Exact Match (%) | 50 |
| HELM Classic - TruthfulQA | 21.66 | Exact Match (%) | 40.9 |
| HELM Classic - LSAT | 18.7 | Exact Match (%) | 25 |
| HELM Classic - MMLU | 26.42 | Exact Match (%) | 22.7 |
| HELM Classic - RAFT | 49.17 | Exact Match (%) | 16.7 |
| HELM Classic - MS MARCO TREC | 30.37 | NDCG@10 (%) | 13.3 |
| HELM Classic - Dyck | 35.8 | Exact Match (%) | 13.2 |
| HELM Classic - BLiMP | 79.3 | Exact Match (%) | 12.9 |
| HELM Classic - Synthetic Reasoning Natural | 0 | F1 (%) | 12.5 |
| HELM Classic - Synthetic Reasoning Abstract | 12.1 | Exact Match (%) | 10.3 |
Interactive version: theaggregate.ai/model?slug=cohere-small-v20220720-410m · How It Works · Data refreshed daily, snapshot 2026-11-08.