OpenAI Compatible API Documentation

Base URL and Authentication

The base URL for all requests is HTTPS://api.keyapikimi.com/v1. Authenticate by passing your unique API key in the Authorization header. You receive this key instantly after signing up with Google or email. The key grants access to the uncensored model without requiring a monthly subscription.

Copycurl https://api.keyapikimi.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "fetch weather in tokyo"}]
  }'

Your API key remains active until you replace it. Errors and refusals do not consume your prepaid credit. Privacy is maintained because prompts are not used for training your data.

First Request: Chat Completions

Send text queries to the single uncensored model via the chat completions endpoint. The model id is "uncensored". This endpoint accepts standard parameters like temperature and top_p. It returns generated text without content refusals for lawful adult use.

Copyfrom openai import OpenAI

client = OpenAI(base_url="https://api.keyapikimi.com/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)

Each request supports a context window of 100,000 tokens for both input and output combined. You can stream responses or receive them as a single block. The model handles function calling and JSON mode efficiently.

Node.js SDK Integration

Use the official OpenAI SDK for Node.js to connect easily. Set the base URL and API key in your configuration object. This approach works with any OpenAI-compatible client library. You avoid complex routing logic by using a single model provider.

Copyimport OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.keyapikimi.com/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);

Pass the model id "uncensored" to ensure requests go to the correct server. The API handles token counting and billing automatically. Errors are returned as standard HTTP status codes for easy debugging.

Streaming Responses via SSE

Enable streaming by setting the stream parameter to true. The API returns Server-Sent Events (SSE) for real-time token delivery. Token usage details appear in the final chunk of the stream. This method reduces perceived latency for long outputs.

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)

Parse the SSE events to display text as it generates. The max output per request is 16,000 tokens. If you do not set max_tokens, the default is 2,048 tokens. Streaming works with all standard parameters like stop sequences.

Parameters: Temperature and Top P

Control creativity with the temperature parameter. Higher values produce more diverse text. Use top_p to limit sampling to the most probable tokens. The stop parameter halts generation when specific strings appear. These settings work together for precise control.

Adjust seed for reproducible results during testing. Presence and frequency penalties help reduce repetition. JSON mode forces the output to be valid JSON. Use response_format to enforce strict data structures for your applications.

Limits, Errors, and Context Window

The context window is 100,000 tokens total. You can send up to 300 requests per minute per key. Concurrency is limited to 8 simultaneous requests per key. The request body must not exceed 8 MB. These limits ensure stable performance for all users.

Handle HTTP 401 for invalid keys. HTTP 402 indicates insufficient prepaid credit. HTTP 429 means you hit the rate limit. Credit never expires, so you can top up anytime via crypto. Refunds are not issued for used credit.