Kimi K2.6 API
New256K context · Released Jul 18, 2026 · By Moonshot AI · Modified MIT
Kimi K2.6 costs $0.55 per million input tokens and $2.3 per million output tokens, against an official rate of $0.95 and $4 — 42% below official pricing. It is a Moonshot AI model under the Modified MIT licence, with a 256K context window, callable through an OpenAI-compatible endpoint with the model id `moonshotai/kimi-k2.6`.
Input
$0.55/M
$0.95/M
Output
$2.30/M
$4.00/M
Savings
42%
off official
- No markup
- OpenAI SDK compatible
- No minimum spend
- Modified MIT
Moonshot's agentic flagship. Holds its reasoning together across very long tool-use chains, which is where most cheaper models quietly fall apart.
Copy this and go
The model id is already filled in. Nothing in this snippet needs editing except your key.
from openai import OpenAI
client = OpenAI(
api_key="YOUR_API_KEY",
base_url="https://router.xark.io/api/v1" # ← the only line that changes
)
response = client.chat.completions.create(
model="moonshotai/kimi-k2.6",
messages=[{"role": "user", "content": "Hello"}]
)
Kimi K2.6 pricing across providers
Rates as published by each provider. We update these when they change; if you spot one that's stale, tell us and we'll fix it.
| Provider | Input | Output |
|---|---|---|
| AI Token Router | $0.55/M | $2.30/M |
| Official (Moonshot AI) | $0.95/M | $4.00/M |
| OpenRouter | $0.92/M | $3.88/M |
| DeepInfra | $0.68/M | $2.75/M |
| Together AI | $0.90/M | $3.60/M |
Cached input on Kimi K2.6 is billed at $0.11/M — roughly 80% below the standard input rate. For agent loops with a large fixed system prompt this is usually the number that matters most.
What it's good at
- Long-horizon coding tasks and multi-file refactors
- Agent workflows that chain dozens of tool calls
- Repo-scale code review and migration passes