Uncensored LLM API Quickstart

Set your base URL and API key

Point your OpenAI SDK to our endpoint and paste your key from the dashboard. You will see your key on the Get API key page immediately after signing in with Google or an email address. The base URL is HTTPS://api.keyapiaiopen.com/v1, which ensures compatibility with any standard OpenAI-compatible client. You do not need to select a model; we route all traffic to our single uncensored LLM. Your key is active instantly, and you can swap it out if you lose it by generating a new one in the settings.

Copycurl https://api.keyapiaiopen.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 errors in this python script"}]
  }'

Keep your base URL consistent across all services. Whether you use a local wrapper or a cloud frontend, this URL handles the routing. We do not offer image or embedding endpoints, so stick to chat-completions. If you are migrating from another provider, just replace the base URL and the key string. The protocol remains identical to the official standard.

Make your first chat completion request

Send a POST request to /v1/chat/completions with the model id uncensored to get your first response. Our API accepts standard parameters like temperature and top_p to control creativity. You can send up to 8 requests simultaneously per key, but do not exceed 300 requests per minute. The model is tuned to answer without refusals for lawful adult content, making it ideal for NSFW chatbot use cases. There are no hidden fees for errors; you only pay for successful token usage. This simplicity contrasts with complex routing systems that often add latency.

Copyfrom openai import OpenAI

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

Use the official SDK for the smoothest experience. Pass your key as an environment variable or directly in the client initialization. The response will contain the generated text in the choices array. If you need specific formatting, use the response_format parameter. The system handles the heavy lifting of model inference on our servers.

Handle streaming responses

Enable streaming by setting stream: true to receive tokens in real-time via Server-Sent Events. This is crucial for applications that need to display text as it is generated. Each chunk contains partial data, and the final chunk includes the total token usage statistics. Streaming reduces perceived latency significantly for long outputs. You can parse the SSE stream in any language that supports HTTP streaming. Our infrastructure ensures that the stream remains stable even under heavy load.

Copyimport OpenAI from "openai";

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

Node.js developers can use the built-in fetch API or libraries like openai to handle streams. Be sure to handle potential network interruptions gracefully. The stream ends when the model finishes generation or when you reach the token limit. You can stop the stream early by closing the connection. This feature is essential for interactive chat interfaces.

Enable JSON mode or tool calling

Use response_format: {"type": "json_object"} to force the model to output valid JSON. This is useful for structured data extraction or API integrations. You can also define tools for function calling to let the model decide which action to take. The model supports tool_choice parameters for precise control. JSON mode is particularly helpful for developers building agents that need to parse complex responses. Error handling is built-in; invalid JSON will be returned as a string if the model fails to comply. Always validate the output in your client code.

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)

Combine streaming with JSON mode for real-time structured data. This approach is powerful for building dynamic dashboards. The model will try to maintain valid JSON syntax throughout the stream. If the output is too long, the stream will end naturally. You can adjust the max_tokens parameter to control the output length. This flexibility makes our API suitable for both casual chat and serious development.

Understand token limits and pricing

Our pricing is transparent: $0.15 per 1M input tokens and $0.50 per 1M output tokens. You pay only for what you use, with no monthly subscription fees. Your prepaid credit never expires, so you can top up whenever you want. We accept crypto only: USDT on TRC20 or USDC on Base. Minimum top-up is $20, and you get bonus credit for larger amounts. Errors and refusals are free, so you do not waste credits on failed requests. The context window is 100,000 tokens, allowing for long documents. Output is limited to 16,000 tokens per request. If you do not set max_tokens, the default is 2,048 tokens. Rate limits are 300 requests per minute. You will receive a 429 error if you exceed this. A 401 error means your key is invalid, and a 402 error means you have no credit. Keep your base URL and key secure to avoid unauthorized usage. This uncensored AI API key gives you direct access without middlemen.

Test with the uncensored model

The model id is uncensored, and it is an open-weight model tuned for adult content. It does not refuse lawful controversial topics or NSFW LLM requests. It is not GPT, Claude, or any other vendor's model. We run it on our own servers to ensure consistent performance. There is a hard limit: sexual content involving minors is always refused. You can test this by sending a prompt with adult themes. The model will respond without the typical guardrails found in other APIs. This makes it perfect for creators who need raw output. You can use any OpenAI-compatible client, including popular frontends like LobeChat or TypingMind. Just update the base URL and key. The uncensored AI API key is your gateway to unrestricted generation. No phone number is required for signup. Sign in with Google or email. Your data is not used for training. This privacy-focused approach appeals to developers who value their input. The trial credit of $0.50 lets you test before you buy. Get your key now and start generating.