Uncensored ai API Reference

Base URL and Authentication

Your API key authenticates every request. Pass it via the Authorization header using the Bearer scheme. This works with any OpenAI-compatible client. You can generate a key instantly after signing up with Google or email. The key is shown once on the dashboard. Keep it secure.

Use this base URL for all endpoints. It routes your requests to our uncensored model infrastructure. Change only the base_url in your SDK configuration. Do not modify the path structure. This ensures compatibility with existing tooling.

Authentication is simple. We do not use complex OAuth flows. Just include the header. If you lose your key, generate a new one. It replaces the old one immediately. One active key per account keeps things simple.

First Request with cURL

Send your first request using standard HTTP tools. This confirms your key works. Replace YOUR_API_KEY with your actual key. The model id is "uncensored". This returns text directly.

Copycurl https://api.aiuncencored.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "List the top 5 python libraries for web scraping"}]
  }'

Check the response for the generated text. If you get a 401, your key is wrong. A 402 means you need to top up. Use crypto to add credit. USDT on TRC20 or USDC on Base are accepted. The trial gives you $0.50 to start.

Python SDK Integration

Use the official OpenAI Python library. Install it via pip. Configure the client with your base URL and key. This approach works for most Python projects. It simplifies the request structure significantly.

Copyfrom openai import OpenAI

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

Call the chat completions endpoint. Pass the messages array. The model handles the rest. You get raw text back. This is ideal for automation scripts. You can process the output programmatically. The SDK handles the JSON serialization. You focus on your logic.

Node.js SDK Usage

Integrate with Node.js using the openai package. Install it in your project. Set the base URL and API key. This mirrors the Python approach. It works with Express or standalone scripts.

Copyimport OpenAI from "openai";

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

Send a chat completion request. The uncensored model responds without standard filters. This is useful for creative writing or data generation. The code is straightforward. Error handling is built into the SDK. You can catch specific HTTP errors. This makes debugging easier. Your application stays robust.

Streaming and SSE Usage

Enable streaming by setting stream to true. This returns tokens as they are generated. It reduces perceived latency. Users see output instantly. This is better for chat interfaces.

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)

Each chunk contains partial text. The final chunk includes token usage data. This helps you track costs accurately. You can update the UI in real time. The uncensored AI API supports this natively. It works with any SSE consumer. This is crucial for a good user experience. Do not forget to handle the end event.

Limits, Errors, and Context

Your requests are limited to 300 per minute. You can run 8 concurrent requests. The request body must be under 8 MB. These limits prevent abuse. They ensure fair usage for all developers. If you exceed the limit, you get a 429 error. Wait and retry.

A 401 error means your key is invalid. Check your spelling. A 402 error means you have no credit. Top up with crypto. The context window is 100,000 tokens. This includes both prompt and completion. Max output is up to the full 100,000-token window. Errors and refusals are free. This protects your balance. Pay only for successful tokens.