Skip to main content

​
Kimi K2.6

Pick this for multi-step reasoning, planning, and tool selection — the structured thinking trace surfaces in the playground so failures are debuggable.
Model ID: moonshotai/Kimi-K2.6

​
About

Kimi K2.6 emits a structured thinking trace before the final answer. Best for multi-step reasoning, planning, and tool selection. The Reasoning accordion in the playground exposes the inline trace as it streams.

​
Pricing

DirectionRate (USD)
Input$0.55 / 1M tokens
Output$2.20 / 1M tokens

​
Capabilities

  • chat
  • reasoning
  • tool-use
  • json-mode

​
Specs

  • Context window: 262k tokens
  • License: Modified MIT
  • Released: March 22, 2026
  • Chat and instruction following
  • Multi-step reasoning and planning
  • Tool use and function calling
  • Structured-output generation (JSON mode)

​
Benchmarks

Benchmarks fill in during Alpha. Tracked as a follow-up to RD-562.

​
Example prompts

  • Plan a migration of a 50M-row Postgres table to a NOT NULL column without downtime.
  • Given this stack trace, identify the most likely root cause and suggest 3 next debugging steps.

​
Quick start

curl https://api.corvex.ai/v1/chat/completions \
  -H "Authorization: Bearer $CORVEX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "moonshotai/Kimi-K2.6",
    "messages": [{"role": "user", "content": "Hello"}]
  }'
Open in the Corvex playground