Kimi API on your Kimi membership

Three ways to get a Kimi API

Kimi sells API access in two ways: pay-as-you-go billing on the Kimi API Platform, and an API key included with Kimi Code on paid memberships. Plan2API sits on top of that second option.

The two Kimi products use different base URLs and different keys. Platform keys from the Kimi API Platform point at https://api.moonshot.ai/v1, while Kimi Code keys start with sk-kimi- and come from the Kimi Code Console for https://api.kimi.ai/coding/v1, so a key from one won't work on the other. Kimi Code's own endpoint already speaks the OpenAI protocol, so if Kimi is the only model you use, it already works with OpenAI SDKs.

Kimi memberships that include Kimi Code

Kimi Code comes with every Kimi membership from Plus up, and switching to annual billing lowers the monthly price. The Free plan doesn't include Kimi Code.

On Kimi Code, K3's 1M-token context and the Highspeed models require Kimi Pro or a higher tier, so Plus members get K3 with a 256K context window. Kimi also warns that K3.1 Pro requests may queue or fail at launch.

Connect Kimi Code in 3 steps

1. Create an API key in the Kimi Code Console at kimi.com/code/console (your key will start with sk-kimi-).
2. Sign up at plan2api.com/connect with a Google account and paste in your Kimi Code key.
3. Copy the p2a_ API key you get after connecting, and grab the copy-paste setup for curl, Python or Node from the dashboard Quickstart page.

Here is the official OpenAI Node SDK calling kimi/k3 at https://api.plan2api.com/v1 with reasoning effort set to max:

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.plan2api.com/v1",
  apiKey: process.env.PLAN2API_KEY,
});

const reply = await client.chat.completions.create({
  model: "kimi/k3",
  reasoning_effort: "max",
  messages: [{ role: "user", content: "Plan a migration from REST to gRPC in five steps." }],
});

console.log(reply.choices[0].message.content);

Kimi models on Plan2API

The table below lines up Kimi Code's K3 series against Kimi API Platform pricing and shows which model Plan2API serves.

• On Plan2API, Kimi K3 is kimi/k3 (or alias moonshotai/kimi-k3) with a 1,048,576-token context and text and image input.
• Each conversation starts on the 256K window and switches to the 1M window once it passes about 192K tokens, and on Kimi Code the 1M version of K3 uses about twice the quota of the 256K version.
• Calling GET /v1/models lists every model id Plan2API serves.

What works and what doesn't

• Requests use the OpenAI Chat Completions format on POST /v1/chat/completions, covered in the API documentation.
• Streaming works.
• Tool and function calling work.
• Reasoning effort works via reasoning_effort or OpenRouter-style reasoning.effort, taking low, high or max on K3 with high as the default.
• Image input works, and Plan2API fetches image URLs in messages for you (up to 32 per request, up to 20 MB and 15 seconds each).
• JSON mode works with response_format set to json_object or json_schema.
• temperature and top_p are fixed on Kimi at 1 and 0.95.
• Plan2API does not have /v1/responses or Anthropic /v1/messages, while Kimi Code's own endpoint offers both.
• The API sends Access-Control-Allow-Origin: * CORS headers, so browser apps can call it directly.

Quota, limits and Kimi's usage rules

Plan2API adds no token quota of its own, so your Kimi membership's quota applies. Kimi Code shares that membership quota with the Kimi Code apps, and all your devices and API keys draw from the same pool. New plans run on a rolling 5-hour rate window plus a monthly total with no weekly limit (legacy plans refresh every 7 days), and the Kimi Code Console shows what you have left.

When your quota runs out, Kimi lets subscribers turn on Extra Usage, which is a prepaid balance priced close to the Kimi API Platform. If you connect two or more Kimi accounts to Plan2API, a request that hits a limit moves to the next account before any output is sent, and when every connected account is out, the API returns HTTP 429 subscription_quota_exhausted with a Retry-After header set to the earliest reset.

One key for Kimi and your other plans

Plan2API makes sense on top of Kimi Code when you want a single LLM gateway with one p2a_ key and one base URL (https://api.plan2api.com/v1) routing to your Kimi, Claude, ChatGPT (Codex) and Gemini subscriptions by model prefix (kimi/ or moonshotai/ for Kimi), plus automatic failover across multiple Kimi accounts.

Your Kimi Code key is encrypted with HPKE (X25519) and AES-256-GCM so only your p2a_ API key can unlock it. The API key's secret is never stored, prompts and outputs are never stored, and Plan2API's servers and database run on OVHcloud in Germany (see how security works).

• Free is $0 with 1 connected subscription, 2 API keys, 50 chat requests a day and no card.
• Pro is $12 a month or $99 a year with 3 connected subscriptions, 10 API keys and no daily request cap.
• Scale is $29 a month or $249 a year with 25 connected subscriptions, 50 API keys and no daily request cap.