Kimicode: First request in five minutes

Base URL and Authentication

Start by pointing your client to our base URL: https://api.kimicode.top/v1. You need an API key for every request. Generate one by clicking Get API key on the site. Use Google sign-in or an email and password. No phone number is required. The key is shown immediately after creation. Keep it secure; there is one active key per account. If you generate a new key, the old one stops working. Pass the key in the Authorization header as a bearer token.

First Request

Send your first prompt to the chat completions endpoint. Use the model ID uncensored. The API accepts standard OpenAI-compatible formats. Input goes in, text comes out. Errors and refusals are free, so you only pay for successful token usage. Test connectivity with a simple prompt.

Copycurl https://api.kimicode.top/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Write a blunt product review of a cheap VPN."}]
  }'

If the key is invalid, you get a 401 error. If you lack credit, you get a 402. Check your account balance if requests fail unexpectedly.

Python SDK Integration

Use the official OpenAI Python SDK. Set the base URL to our endpoint. Pass your API key. The client handles the request format automatically. This works for any language that supports OpenAI-compatible clients. Install the library via pip. Configure the client with your base URL and key. Send a message using the chat completions endpoint. The response contains the generated text. You can parse the content directly from the response object.

Copyfrom openai import OpenAI

client = OpenAI(base_url="https://api.kimicode.top/v1", api_key="YOUR_KEY")

resp = client.chat.completions.create(
    model="uncensored",
    messages=[{"role": "user", "content": "Summarise this thread without softening it."}],
)
print(resp.choices[0].message.content)

Ensure your code handles exceptions gracefully. Network issues or rate limits may require retries.

Node SDK Usage

The Node.js SDK works identically. Import the client. Configure the base URL and API key. Call the chat completions method. Pass the model ID and messages array. The response structure matches the OpenAI format. Extract the assistant message content. This approach works for server-side generation or edge functions. No additional configuration is needed beyond the base URL change.

Copyimport OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.kimicode.top/v1", apiKey: process.env.API_KEY });

const resp = await client.chat.completions.create({
  model: "uncensored",
  messages: [{ role: "user", content: "Draft a villain monologue for my game." }],
});
console.log(resp.choices[0].message.content);

Handle errors by checking the status code. Rate limits return 429. Insufficient funds return 402.

Streaming Responses

Enable streaming by setting stream: true. The API returns Server-Sent Events (SSE). Each chunk contains partial content. Token usage details appear in the last chunk. This reduces perceived latency for long outputs. Handle stream termination properly. Close the connection when the stream ends or an error occurs. Streaming works with all standard clients that support SSE.

Copystream = client.chat.completions.create(
    model="uncensored",
    messages=[{"role": "user", "content": "Tell the story in second person."}],
    stream=True,
)
for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

Ensure your client buffers chunks correctly. Concatenate the content to form the final response.

Limits, Errors, and Context

Context window is 100,000 tokens total. Max output is 32,000 tokens per request. Set 2,048 if max_tokens is unset. Rate limit is 300 requests per minute per key. Concurrency limit is 8 simultaneous requests. Request body max is 8 MB. Errors include 401 for invalid keys, 402 for no credit, and 429 for rate limits. Credit never expires. Errors are free. No subscription traps.