New Finance Agent Benchmark Released

Released date

Model

Last Updated 5/7/2025

mistralai/mistral-medium-2505

Mistral Medium 3 (05/2025)

Mistral Medium 2505

Released Date: 5/7/2025

Avg. Accuracy:

68.7%

Latency:

14.37s

Performance by Benchmark

Benchmarks

Accuracy

Rankings

58.1%

( 18 / 38 )

58.1%

18 / 38

84.9%

( 6 / 59 )

84.9%

6 / 59

66.2%

( 33 / 66 )

66.2%

33 / 66

73.4%

( 22 / 46 )

73.4%

22 / 46

60.5%

( 22 / 26 )

60.5%

22 / 26

87.0%

( 17 / 42 )

87.0%

17 / 42

42.3%

( 14 / 36 )

42.3%

14 / 36

91.6%

( 11 / 40 )

91.6%

11 / 40

61.3%

( 53 / 64 )

61.3%

53 / 64

78.2%

( 27 / 44 )

78.2%

27 / 44

52.5%

( 22 / 38 )

52.5%

22 / 38

74.4%

( 20 / 37 )

74.4%

20 / 37

63.1%

( 17 / 23 )

63.1%

17 / 23

Academic Benchmarks

Proprietary Benchmarks (contact us to get access)

Cost Analysis

Input Cost

$0.40 / M Tokens

Output Cost

$2.00 / M Tokens

Input Cost (per char)

$0.07 / M chars

Output Cost (per char)

$0.40 / M chars

Join our mailing list to receive benchmark updates on

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