Kimi K3

Kimi’s flagship model for long-horizon coding and end-to-end knowledge work, with a 1M-token context window.

ReasoningTool UseImplicit CachingFile InputVision (Image)

index.ts

import { streamText } from 'ai'

const result = streamText({
  model: 'moonshotai/kimi-k3',
  prompt: 'Why is the sky blue?'
})

Throughput24 hours

P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.