Kimi K3
moonshotai/kimi-k3Moonshot's flagship multimodal model, with always-on reasoning and a selectable effort level.
- Text, Image, Video
- $6 / $30
- 1M
- Jul 22, 2026
Overview
Moonshot's flagship multimodal model, with always-on reasoning and a selectable effort level. It is billed from the same balance as every other model and API here, and the price of each call comes back in the response.
Run Kimi K3 with an agent
Each guide pairs this model with a currently compatible agent, explains their separate roles, and opens the shared Composer with both selections ready.
- Open guide
Hermes + Kimi K3
Compatibility, setup, current model facts, pricing boundaries, and a preselected Composer.
- Open guide
OpenClaw + Kimi K3
Compatibility, setup, current model facts, pricing boundaries, and a preselected Composer.
- Open guide
pi + Kimi K3
Compatibility, setup, current model facts, pricing boundaries, and a preselected Composer.
- Open guide
DeepSeek Harness + Kimi K3
Compatibility, setup, current model facts, pricing boundaries, and a preselected Composer.
- Open guide
Kimi Code + Kimi K3
Compatibility, setup, current model facts, pricing boundaries, and a preselected Composer.
- Open guide
OpenCode + Kimi K3
Compatibility, setup, current model facts, pricing boundaries, and a preselected Composer.
Pricing
The default plan, in USD per million tokens.
| Meter | Per 1M tokens |
|---|---|
| Input | $6.00 |
| Output | $30.00 |
| Cache read | $0.6 |
| Cache write | $6.00 |
Your balance and per-call records come from /v1/credits and /v1/generation.
Use it
Point your client at the gateway and pass this model id. Nothing else changes.
import Anthropic from "@anthropic-ai/sdk";
const client = new Anthropic({
baseURL: "https://gateway.agentsky.dev",
apiKey: process.env.AGENTSKY_API_KEY, // ast_…
});
const message = await client.messages.create({
model: "moonshotai/kimi-k3",
max_tokens: 256,
messages: [{ role: "user", content: "Summarise this ticket." }],
});
console.log(message.usage.cost); // USD, in the response