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:

49.1%

Latency:

17.31s

Performance by Benchmark

Benchmarks

Accuracy

Rankings

FinanceAgent

4.3%

( 30 / 34 )

CorpFin

54.5%

( 30 / 45 )

TaxEval

66.6%

( 45 / 62 )

Math500

76.2%

( 38 / 54 )

AIME

13.3%

( 42 / 52 )

MGSM

85.7%

( 47 / 55 )

LegalBench

78.7%

( 30 / 77 )

MedQA

80.5%

( 40 / 58 )

GPQA

29.3%

( 51 / 54 )

MMLU Pro

65.2%

( 45 / 52 )

LiveCodeBench

35.1%

( 48 / 54 )

SWE-bench

0.2%

( 16 / 16 )

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)

$3.33 / M chars

Join our mailing list to receive benchmark updates on

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