New Finance Agent Benchmark Released

Released date

Model

Last Updated 7/11/2025

together/moonshotai/Kimi-K2-Instruct

Kimi K2 Instruct

Kimi K2 Instruct

Released Date: 7/11/2025

Avg. Accuracy:

74.8%

Latency:

32.78s

Performance by Benchmark

Benchmarks

Accuracy

Rankings

62.2%

( 14 / 41 )

62.2%

14 / 41

72.0%

( 50 / 64 )

72.0%

50 / 64

66.4%

( 34 / 70 )

66.4%

34 / 70

74.1%

( 21 / 51 )

74.1%

21 / 51

94.2%

( 7 / 47 )

94.2%

7 / 47

62.7%

( 12 / 41 )

62.7%

12 / 41

90.9%

( 17 / 45 )

90.9%

17 / 45

84.0%

( 25 / 49 )

84.0%

25 / 49

71.5%

( 13 / 42 )

71.5%

13 / 42

70.4%

( 7 / 42 )

70.4%

7 / 42

Academic Benchmarks

Proprietary Benchmarks (contact us to get access)

Cost Analysis

Input Cost

$1.00 / M Tokens

Output Cost

$3.00 / M Tokens

Input Cost (per char)

$0.28 / M chars

Output Cost (per char)

$0.83 / M chars

Join our mailing list to receive benchmark updates on

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