API documentation for uncensored inference

Base URL and Authentication

Your requests target https://api.cursorkey.cc/v1. You must include your API key in the Authorization header. The key is displayed immediately after signup. It is unique to your account. One key replaces the previous one.

Use the official SDKs. Change the base URL. Pass the key. That is all.

Copycurl https://api.cursorkey.cc/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "List available cursor key bindings for VS Code"}]
  }'

First Request

Send a standard chat completion. The model ID is uncensored. The system ignores refusals for lawful adult content. You get raw text. The context window holds 100,000 tokens total. This includes both input and output. Keep your payloads under 8 MB.

Copyfrom openai import OpenAI

client = OpenAI(base_url="https://api.cursorkey.cc/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)

Node SDK Integration

Install the official package. Point it at our base URL. The behavior matches the standard OpenAI interface. You can stream responses. You can use JSON mode. The client handles the protocol. You handle the logic. Errors return standard HTTP codes.

Copyimport OpenAI from "openai";

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

Streaming Responses

Enable streaming by setting stream: true. Tokens arrive sequentially. The final chunk contains token usage statistics. This allows you to track costs in real time. The server pushes data via Server-Sent Events. Your client consumes the stream.

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)

Function Calling and JSON Mode

Define tools in your request. The model returns structured data. Use response_format: {"type": "json_object"} for strict JSON. This ensures predictable parsing. The model supports tool choices. You can force specific functions. This works reliably with our uncensored model.

Rate Limits and Constraints

Limit yourself to 300 requests per minute. You may run 8 concurrent requests. A 401 error means your key is invalid. A 402 error indicates no credit. A 429 error means you hit the rate limit. Refusals are free. Errors are free. Only completed tokens count. The model refuses minor content only.