New Finance Agent Benchmark Released

An update to Google's Gemini 2.5 Flash model, with improved performance on several benchmarks.

Released Date: 9/25/2025

Avg. Accuracy:

67.1%

Latency:

61.27s

Performance by Benchmark

Benchmarks

Accuracy

Rankings

FinanceAgent

30.6%

( 17 / 39 )

CorpFin

67.8%

( 12 / 52 )

CaseLaw

68.2%

( 20 / 35 )

TaxEval

67.5%

( 47 / 69 )

MortgageTax

68.9%

( 23 / 41 )

AIME

51.4%

( 26 / 59 )

MGSM

89.8%

( 36 / 62 )

LegalBench

81.7%

( 14 / 84 )

MedQA

91.2%

( 20 / 65 )

GPQA

75.8%

( 11 / 61 )

MMLU Pro

83.7%

( 11 / 59 )

MMMU

80.8%

( 3 / 38 )

Terminal-Bench

26.3%

( 13 / 20 )

SWE-bench

55.6%

( 7 / 21 )

Academic Benchmarks
Proprietary Benchmarks (contact us to get access)

Cost Analysis

Input Cost

$0.30 / M Tokens

Output Cost

$2.50 / M Tokens

Input Cost (per char)

N/A

Output Cost (per char)

$0.76 / M chars

Join our mailing list to receive benchmark updates on

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