Command-R+ (Apr '24): benchmark results

Provider: Cohere. Released 2024-04-04. Access: Open.

Unified ELO 1474 ± 59, rank #818 of 1605 rated models, from 19 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Enkrypt AI - Jailbreak Risk9.28Risk Score49.7
PLCC - Culture & Tradition52Accuracy (%)30.8
PLCC - Vocabulary46Accuracy (%)30.6
PLCC - Art & Entertainment39Accuracy (%)24.4
AA Humanity's Last Exam4.63Accuracy (%)23.7
PLCC - Overall49.33Mean category accuracy (%)23.7
PLCC - History61Accuracy (%)23.3
PLCC - Geography53Accuracy (%)19.8
PLCC - Grammar45Accuracy (%)16.4
Enkrypt AI - Toxicity Risk12.27Risk Score12.3
AA LiveCodeBench12.17Pass@1 (%)11.4
AA MMLU-Pro43.18Accuracy (%)10.1

Interactive version: theaggregate.ai/model?slug=command-r-plus-apr-24 · How It Works · Data refreshed daily, snapshot 2026-09-26.