AI Uncensored: How to Use the uncensored API

What is AI Uncensored?

AI uncesored services remove the standard content filters that restrict topic selection in commercial models. This allows for unrestricted generation of adult themes, controversial opinions, and niche creative content without unexpected refusals. Our API provides a hosted environment where you send text and receive text back, focusing purely on language capabilities.

The underlying model is an open-weight LLM tuned to answer without hesitation for lawful adult use. It does not use your prompts for training data, ensuring privacy for your creative projects. You get a single model ID called 'uncensored' that behaves consistently across requests.

Unlike multi-vendor gateways, we do not route your requests between different providers. You interact directly with our infrastructure. This simplicity reduces latency and makes debugging easier when your content gets flagged or accepted.

Why Choose an uncensored API?

An uncensored API gives you predictable behavior for creative writing and roleplay without hidden content blocks. Standard models often refuse NSFW topics even when they are relevant to the story context. Our model is designed to handle these scenarios naturally.

• No Subscription Fees: You pay only for what you use. Credits never expire.
• Transparent Pricing: Costs are calculated per token. There are no hidden base fees.
• Crypto Payments: Top up with USDT or USDC for quick processing.

This approach suits developers building chatbots or story generators. You avoid the frustration of generic filters blocking valid creative output. The system is built for utility, not marketing hype.

Setting Up Your First Request

You can start sending requests immediately after generating your API key. The endpoint is standard and compatible with most existing tools. Use the base URL HTTPS://api.uncesoredai.com/v1 and set your model to 'uncensored'.

Authentication relies on a single API key per account. You can generate this key via Google login or email. The key appears instantly on your dashboard. You do not need to verify a phone number.

Send a POST request to /v1/chat/completions. Include your system prompt to define the persona. The model will respond with text that adheres to your instructions. Errors do not consume your credits.

Example Request

Ensure your JSON structure matches the OpenAI format. This ensures compatibility with libraries like OpenAI Python or Node.js.

Copycurl https://api.uncesoredai.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "List the top five open source LLMs for local deployment"}]
  }'

Configuring for NSFW Content

NSFW content generation works by default. The model does not refuse lawful adult themes. You do not need special parameters to enable this mode. Just send your prompt and receive the response.

However, keep in mind that sexual content involving minors is always refused. This is a hard limit applied to all requests. Everything else is fair game for creative freedom.

You can adjust the temperature to control creativity. Higher values produce more diverse outputs. Lower values are better for consistent storytelling. The model supports parameters like top_p and seed for reproducibility.

Use the stop parameter to prevent the model from rambling. This helps keep your token usage in check. You can also set a max_tokens limit to control the response length.

Handling Tokens and Costs

Costs are calculated based on token usage. Input tokens cost $0.15 per million. Output tokens cost $0.50 per million. You can see the token count in the response headers.

The context window supports up to 100,000 tokens. This includes both the prompt and the completion. You can generate up to 16,000 tokens in a single response.

• Prepaid Credit: Funds are charged by real usage.
• No Expiration: Your balance remains active indefinitely.
• Bonus Credits: Top-ups of $100 or more include extra credit.

Errors are free. If the server returns an error, you are not charged. This protects your budget during testing phases.

Streaming for Real-Time Generation

Streaming allows you to display text as it generates. This improves the user experience for chat applications. The API supports Server-Sent Events (SSE) for streaming.

Enable streaming by setting the stream parameter to true. The response will come in chunks. Each chunk contains part of the generated text.

You can parse the chunks to update the UI in real time. The final chunk includes the total token usage. This helps you track costs accurately.

Streaming reduces the perceived latency for users. It makes the API feel more responsive. You can handle partial responses gracefully even if the connection drops.

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)

Common Use Cases

Developers use this API for various purposes. Roleplay bots benefit from the lack of content filters. Fanfiction writers can generate scenes without interruptions.

Content creators use the model for brainstorming. The high context window allows for long-form writing. You can maintain continuity across chapters.

Testers use the API for security research. The model can discuss controversial topics openly. This is useful for training other models or evaluating bias.

Developers integrating with existing tools find the OpenAI compatibility valuable. They can swap in this API without changing their codebase. The standard endpoints simplify the integration process.

Limitations and Content Policies

The API has specific limits to ensure stability. You can send 300 requests per minute. Each key can handle 8 concurrent requests.

The request body size is limited to 8 MB. This restricts the size of your input data. Large files may need to be chunked.

• Content Limit: No sexual content involving minors.
• Rate Limits: 300 RPM per key.
• Concurrency: 8 simultaneous requests.

These limits are standard for high-throughput APIs. They prevent resource exhaustion. You can monitor your usage in the dashboard.