New Finance Agent Benchmark Released

Kimi K2 0905 is an updated version of Kimi K2, a state-of-the-art mixture-of-experts (MoE) language model with 32 billion activated parameters and 1 trillion total parameters. Kimi K2 0905 has improved coding abilities, a longer context window, and agentic tool use, and a longer (262K) context window.

Released Date: 9/4/2025

Avg. Accuracy:

37.5%

Latency:

149.81s

Performance by Benchmark

Benchmarks

Accuracy

Rankings

Terminal-Bench

37.5%

( 7 / 15 )

Academic Benchmarks
Proprietary Benchmarks (contact us to get access)

Cost Analysis

Input Cost

$0.60 / M Tokens

Output Cost

$2.50 / M Tokens

Input Cost (per char)

N/A

Output Cost (per char)

N/A

Join our mailing list to receive benchmark updates on

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