kimi-k3
Moonshot
A multimodal reasoning model for complex coding, knowledge work, and long-running agent tasks. It works across large codebases, calls tools, and debugs, using images, logs, and test output to refine its output.
- 上下文
- 1M
- 输入 / 1M tokens
- $3
- 输出 / 1M tokens
- $15
- 缓存读 / 1M
- $0.3
- 缓存写 / 1M
- $0