Skip to content

Kimi Code Bench v2

A Moonshot AI internal coding-agent benchmark for realistic software-engineering tasks across mainstream programming languages and production technology stacks.

2 models

Higher is better.

  1. 1Kimi K3Moonshot AI72.9
  2. 2Kimi K2.7 CodeMoonshot AI62.0
Higher is better. Bar colour: blue proprietary, orange open weights, green other.

About this benchmark

Metric
Coding-agent pass rate
Tasks
Realistic coding-agent tasks
Authors
Moonshot AI (2026)

Scores are as reported by model developers or evaluators and may use different settings (for example reasoning effort or tools). Compare with care.