New Finance Agent Benchmark Released
cohere/command-a-03-2025 Command A

Command A is Cohere's most efficient and performant model to date, specializing in agentic AI, multilingual, and human evaluations for real-life use cases.

Released Date: 3/13/2025

Avg. Accuracy:

52.7%

Latency:

15.37s

Performance by Benchmark

Benchmarks

Accuracy

Rankings

FinanceAgent

4.3%

( 25 / 29 )

CorpFin

54.5%

( 25 / 40 )

CaseLaw

79.7%

( 30 / 63 )

ContractLaw

67.7%

( 27 / 69 )

TaxEval

66.6%

( 34 / 50 )

Math500

76.2%

( 31 / 46 )

AIME

13.3%

( 31 / 40 )

MGSM

86.8%

( 33 / 44 )

LegalBench

78.7%

( 25 / 68 )

MedQA

80.5%

( 30 / 48 )

GPQA

29.3%

( 38 / 41 )

MMLU Pro

65.2%

( 34 / 41 )

LiveCodeBench

35.1%

( 35 / 41 )

SWE-bench

0.2%

( 12 / 12 )

Academic Benchmarks
Proprietary Benchmarks (contact us to get access)

Cost Analysis

Input Cost

$2.50 / M Tokens

Output Cost

$10.00 / M Tokens

Input Cost (per char)

N/A

Output Cost (per char)

$3.15 / M chars

Join our mailing list to receive benchmark updates on

Stay up to date as new benchmarks and models are released.