Cohere xlarge v20221108 (52.4B): benchmark results
Provider: Other.
Aggregate Rating 1281 ± 35, rank #8977 of 10916 rated models, from 32 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM Classic - OpenbookQA | 58.8 | Exact Match (%) | 93.5 |
| HELM Classic - XSUM | 15.31 | ROUGE-2 (%) | 90.2 |
| HELM Classic - HellaSwag | 81 | Exact Match (%) | 87.1 |
| HELM Classic - CNN/DailyMail | 15.33 | ROUGE-2 (%) | 85.4 |
| HELM Classic - WikiFact | 34.22 | Exact Match (%) | 83.3 |
| HELM Classic - BLiMP | 83.17 | Exact Match (%) | 80.6 |
| HELM Classic - MS MARCO TREC | 54.99 | NDCG@10 (%) | 80 |
| HELM Classic - IMDB | 95.6 | Exact Match (%) | 79.5 |
| HELM Classic - MS MARCO Regular | 31.5 | RR@10 (%) | 79.3 |
| HELM Classic - NaturalQuestions Closed Book | 36.1 | F1 (%) | 78.8 |
| RealToxicityPrompts | 0.01 | Toxic fraction | 78 |
| HELM Classic - BBQ | 39.73 | Exact Match (%) | 73.2 |
Interactive version: theaggregate.ai/model?slug=cohere-xlarge-v20221108-52-4b · How It Works · Data refreshed daily, snapshot 2026-11-08.