Command R: benchmark results

Cohere Command R model optimized for retrieval-augmented generation and enterprise chat. Provider: Cohere. Released 2024-03-11. Access: Open.

Unified ELO 1416 ± 1, rank #1083 of 1392 rated models, from 74 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
MERA - SimpleAr99.4EM (%)74
MERA - ruDetox30.14Joint Score (%)73.1
MERA - PARus89.2Accuracy (%)71.1
MERA - MultiQ50.58F1 (%)68.9
MERA - CheGeKa29.75F1 (%)67.5
RULER88.9Avg Accuracy (%)61.3
CanAiCode98.9Junior-v2 Python Pass Rate (%)60.2
MERA - ruHateSpeech78.11Accuracy (%)60.1
MERA - ruTiE74.96Accuracy (%)55
HELM NaturalQuestions (Open)72.05F1 (%)53.3
HELM NarrativeQA74.17F1 (%)50
IFEval Leaderboard69.71Final Score50

Interactive version: theaggregate.ai/model?slug=command-r · How It Works · Data refreshed daily, snapshot 2026-09-05.