Uncensored AI API Quickstart

Authentication & Base URL

The uncensored AI online endpoint is your gateway to raw text generation. Use the base URL https://api.uncensoredaionline.com/v1 for all requests. You need an API key, which you get by clicking 'Get API key' on the dashboard. Sign in with Google or an email password; no phone number is required. The key appears immediately. Use it in the Authorization header as a Bearer token. There is only one active key per account. If you generate a new one, the old key stops working. We do not offer multiple keys per account.

Making Your First Request

Send a POST request to /v1/chat/completions with your message. The model ID is uncensored. It is an open-weight model tuned for no refusals on lawful adult content. It is not GPT, Claude, or Llama. You get text in, text out. No images, audio, or embeddings. Use the standard OpenAI format. Our API supports the same structure. You can switch your base URL in any SDK to use this service. Here is a basic curl example to get started.

Copycurl https://api.uncensoredaionline.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "List the top 3 errors in this log file."}]
  }'

Check the response for the generated text. If you get a 401 error, your key is invalid. A 402 error means you have no prepaid credit. A 429 error means you hit the rate limit of 300 requests per minute. Errors and refusals are free, so you only pay for actual tokens used.

Python SDK Integration

Use the official OpenAI Python package for easy integration. Set the base_url and api_key before calling the client. This allows you to use the same code structure you know. The model ID remains uncensored. You can pass parameters like temperature and top_p to control creativity. The API is fully compatible with the OpenAI SDK interface. This makes switching from other providers straightforward. Install the package and configure the client. Then send your messages just like you would with GPT-4.

Copyfrom openai import OpenAI

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

This approach works for both simple prompts and complex function calling. You can stream responses or get the full text at once. The choice is yours. The backend handles the heavy lifting. You just need to manage your prepaid credit via crypto top-ups.

Node SDK Configuration

The Node.js SDK follows the same pattern. Initialize the client with your base URL and key. The endpoint structure is identical to the Python version. This consistency reduces development time. You can handle errors and rate limits programmatically. The 8 MB request body limit applies here too. If you exceed it, the request will fail. Streaming is supported via SSE. You can process tokens as they arrive. This is useful for long outputs or real-time displays. The model does not refuse content, so your logic should handle the output as needed.

Copyimport OpenAI from "openai";

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

Remember that credit never expires. You pay per token. Input tokens cost $0.15 per million. Output tokens cost $0.50 per million. There are no hidden fees. The prepaid model means you control your costs. No subscription lock-in. Just buy crypto credit and use it.

Streaming Responses & Token Usage

Enable streaming by setting stream to true. You receive chunks of text as they are generated. The last chunk contains the token usage statistics. This allows you to track costs in real time. Streaming is essential for good user experience in chat interfaces. The uncensored LLM supports this feature fully. You can handle partial responses gracefully. Errors during streaming still count as free if no tokens were generated. Check the usage field in the final chunk. It shows prompt tokens and completion tokens. This data helps you budget your crypto top-ups.

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)

The context window is 100,000 tokens total. This includes both the prompt and the completion. You can output up to 16,000 tokens per request. If you do not set max_tokens, it defaults to 2,048. Plan your prompts accordingly. Long conversations may require summarization to stay within limits. This is a trade-off for the low cost and high capacity.

Function Calling, JSON Mode & Limits

Use function calling for structured data extraction. Define your tools in the request. The model will return JSON matching your schema. You can also force JSON mode with response_format: {"type": "json_object"}. This is useful for parsing data or generating configs. The API supports temperature, top_p, stop, and seed parameters. Adjust these to fine-tune the output. The hard limit is no sexual content involving minors. Everything else is fair game. The 300 requests per minute limit is generous. The 8 concurrent requests limit prevents spam. Use these limits wisely.