BenchLeader
Menu
LeaderboardsModelsProvidersBenchmarksCompareReleasesBlogMethodology
CohereAuto-detected

Command R+

Best configuration ranks #597 of 610 on the BenchLeader Index at 35.0 ±11.0. Last measured 2 Sept 2026.

Blended price
–
Output speed
–
First answer
–
Context
128k
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Overall index35
  2. Reasoning40
  3. Coding18
  4. Knowledge12
  5. Human preference42
  6. Composite36

Benchmark results

One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

BenchmarkdefaultSource
LMArena Hard Prompts1251#266LMArena
GPQA Diamond (AA)not in index32.3%#480Artificial Analysis
Humanity's Last Exam (AA)not in index4.6%#406Artificial Analysis
GPQA Diamond (Vals)not in index31.1%#127Vals AI

Coding

BenchmarkdefaultSource
LMArena Coding1273#270LMArena
LiveCodeBench18.2%#134Vals AI

Knowledge

BenchmarkdefaultSource
MMLU-Pro44.0%#128Vals AI
LegalBench68.3%#119Vals AI
MedQA2.6%#90Vals AI

Human preference

BenchmarkdefaultSource
LMArena Text1261#260LMArena

Composite

BenchmarkdefaultSource
Epoch Capabilities Indexnot in index119.3#165Epoch AI Benchmarking Hub
AA Intelligence Index5.3#496Artificial Analysis

See also

Best for codingAll Cohere models

Data as of 9 Sept 2026. Compare with another model.