Noometry

Model comparison

Command R vs GPT-4

Command R is the stronger model overall, scoring 31.0 to 28.9 on the Noometry Index.

Last verified . 22 shared benchmarks.

Command R Cohere

31.0

Rank #267 Confirmed

GPT-4 OpenAI

28.9

Rank #308 Confirmed

Summary

  • They share 22 benchmarks with published results for both. Command R scores higher in 3 categories and GPT-4 in 5 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Command R leads 25.8 to 10.1.
  • The biggest single-benchmark swing is DTBench: 46.4% for Command R and 62.7% for GPT-4.
  • Command R is cheaper at $0.15 / $0.60 per million input/output tokens, against $30 / $60 for GPT-4.
  • Command R accepts more context: 128K tokens versus 8K.
  • Command R has downloadable open weights; the other is API-only.

Side by side

Command RGPT-4
ProviderCohereOpenAI
Noometry Index31.028.9
Released2024-08-302023-03-14
WeightsOpenProprietary
Context window128K8K
Max output4K8K
Input $ / M tokens$0.15$30
Output $ / M tokens$0.60$60
Results tracked2938

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding GPT-4 leads

Command R: 28.8 (#301), GPT-4: 31.2 (#277)

BenchmarkCommand RGPT-4
BigCodeBench Instruct37.1%46%
LMArena Coding11701254
BigCodeBench Complete45.2%57.2%
WeirdML—12.4%
LiveBench Coding17.9%—
HumanEval+—79.3%

Agentic & Tool Use Not comparable

Command R: —, GPT-4: —

BenchmarkCommand RGPT-4
METR Time Horizons—36.1%

Reasoning GPT-4 leads

Command R: 13.3 (#329), GPT-4: 17.3 (#285)

BenchmarkCommand RGPT-4
LMArena Hard Prompts11641241
DTBench46.4%62.7%
LMCA9.2%17.1%
Chess Puzzles—4%
LiveBench Reasoning21.9%—
Mystery Game Puzzles—12%
LiveBench Data Analysis33.3%—
BIG-Bench Hard—75.1%
Epoch Capabilities Index—125.88
ForecastBench—57.8
HellaSwag—95.3%
LiveBench27.5%—
WinoGrande—87.5%

Math Command R leads

Command R: 25.8 (#244), GPT-4: 10.1 (#304)

BenchmarkCommand RGPT-4
LMArena Math11551269
OTIS Mock AIME 2024-2025—1.1%
LiveBench Math19.4%—
MATH Level 5—23%
GSM8K—92%

Knowledge Command R leads

Command R: 30.5 (#219), GPT-4: 18.4 (#277)

BenchmarkCommand RGPT-4
LMArena Expert11381211
MMLU65.2%86.4%
GPQA Diamond—35.7%
TriviaQA—84.8%

Multilingual GPT-4 leads

Command R: 35.9 (#240), GPT-4: 40.8 (#210)

BenchmarkCommand RGPT-4
LMArena Non-English11741246
LMArena Chinese11821242
LMArena French11611282
LMArena German11761251
LMArena Japanese11431209
LMArena Korean11621184
LMArena Russian11741251
LMArena Spanish11501261

Instruction Following GPT-4 leads

Command R: 58.3 (#256), GPT-4: 65.5 (#217)

BenchmarkCommand RGPT-4
LMArena Instruction Following11671241
LiveBench Instruction Following55.6%—

Long Context GPT-4 leads

Command R: 36.3 (#226), GPT-4: 37.7 (#207)

BenchmarkCommand RGPT-4
LMArena Longer Query11981244

Writing & Preference Command R leads

Command R: 38.4 (#248), GPT-4: 35.0 (#263)

BenchmarkCommand RGPT-4
LMArena Text11871263
LMArena Creative Writing11701244
LMArena Multi-Turn11631257
EQ-Bench Creative Writing—752
LiveBench Language16.7%—

Frequently asked questions

Is Command R better than GPT-4?

Command R is the stronger model overall, scoring 31.0 to 28.9 on the Noometry Index.

Which is cheaper, Command R or GPT-4?

Command R is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GPT-4 lists at $30 and $60.

Is Command R or GPT-4 better for coding?

GPT-4 scores higher on coding benchmarks: 31.2 versus 28.8 in the Noometry coding category.

Which has the bigger context window?

Command R does, with 128K tokens against 8K.

How many benchmarks do Command R and GPT-4 share?

22 benchmarks have published results for both models. Command R has 29 scored results on Noometry and GPT-4 has 38.

Related comparisons

Go deeper