Quickstart for Xtts pipelines

Authentication & Base URL

Start by navigating to the Get API key page and signing in with Google or an email password. You will receive a unique API key immediately. Use the base URL https://api.xtts.cc/v1 for all requests. This URL is compatible with official OpenAI SDKs and any client that supports the OpenAI chat-completions format.

Include your key in the Authorization header as a bearer token. The model identifier you must send is uncensored. This open-weight model is hosted on our servers and is tuned to answer without content refusals for lawful adult use, ensuring your voice pipeline receives the raw text it needs.

Generate Uncensored Text

Send a POST request to /v1/chat/completions with your system and user prompts. The API returns structured JSON containing the generated text. This endpoint is designed for prompt writing, script generation, and post-processing tasks in your voice cloning workflow.

The request body includes the model name, your message history, and optional parameters like temperature or top_p. You can control the output length with the max_tokens parameter. If you do not set max_tokens, the response can run to the end of the window.

Copycurl https://api.xtts.cc/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Write a blunt product review of a cheap VPN."}]
  }'

Python SDK Integration

Use the official OpenAI Python SDK to interact with our API. Set the base_url to our endpoint and provide your API key. This approach allows you to leverage existing code structures while benefiting from our uncensored model's behavior.

Create a client instance with the base URL and then call chat.completions.create. Pass the uncensored model ID and your messages. The response object contains the generated text in content. This method is ideal for batch processing scripts or integrating text generation into larger Python applications.

Copyfrom openai import OpenAI

client = OpenAI(base_url="https://api.xtts.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

For JavaScript environments, use the OpenAI Node SDK. Configure the client with our base URL and API key. This setup enables you to generate text directly in your web or server-side applications.

Call the chat completions endpoint with the uncensored model. The response provides the generated text, which you can then pass to your voice synthesis engine. This integration is suitable for real-time text generation in web apps or backend services that prepare scripts for voice cloning.

Copyimport OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.xtts.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 to true in your request. The API returns Server-Sent Events (SSE) for each generated token. This feature is useful for real-time text display or progressive voice synthesis preparation.

Each event chunk contains partial text. The final chunk includes usage statistics, such as token counts. Streaming ensures your application can process text as it is generated, reducing perceived latency for users. Handle the stream events to accumulate the complete response.

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)

Rate Limits & Parameters

Each API key is limited to 300 requests per minute and 8 concurrent requests. The request body must not exceed 8 MB. If you send an invalid key, you will receive a 401 error. If your prepaid credit is exhausted, you will receive a 402 error. Exceeding the rate limit results in a 429 error.

The context window supports 100,000 tokens total, including both the prompt and the completion. There is no separate output cap, and you can specify a lower max_tokens if you want a shorter reply. Errors and refusals do not consume your prepaid credit. Support for function calling and JSON mode is available for structured data extraction.