Command-R+ — benchmark results

Cohere Command R+ model, a higher-capacity Command R variant for retrieval, tool use, and complex enterprise tasks. Provider: Cohere. Released 2024-04-04. Access: Open.

Unified ELO 1436 ± 16, rank #1064 of 1776 rated models, from 77 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LiveBench Story Generation80.83Score98.6
LiveBench Zebra Puzzle46Score97.9
LiveBench Simplify72.63Score87.5
LiveBench Paraphrase70.22Score86.1
HELM Safety SimpleSafetyTests100LM Evaluated Safety score (%)86
Ru Arena Hard77.17Win Rate (%)71.9
IFEval Leaderboard77.01Final Score70
RABBITS98.28B4BQA Score (%)68.2
BenchBench61.83Aggregate Score (%)66.9
MixEval51.4Score64.7
HELM WMT 201420.33BLEU-4 (%)62.8
LiveBench Summarize62.35Score62.5

Interactive version: theaggregate.ai/model?slug=command-r-plus · How the rankings work · Data refreshed daily, snapshot 2026-07-22.