Uncensored API Chat Completions Documentation

Authentication And Base URL

Start by pasting your key into the Authorization header. We use a standard Bearer token scheme that works with any OpenAI-compatible client. Sign in via Google or email to get your key immediately. No phone number is required for the account setup.

The base URL is HTTPS://api.chataiunfiltered.com/v1. Your requests must point here. The model ID to use is uncensored. This ensures you get the specific uncensored model we host on our own servers.

Copycurl https://api.chataiunfiltered.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "What is the capital of France?"}]
  }'

The Chat Completions Endpoint

Send a POST request to /v1/chat/completions with your prompt. The API returns text directly without extra fluff. You can set parameters like temperature or top_p to control randomness.

Use the Python SDK to integrate quickly. Just set the base URL to our endpoint. The client handles the JSON serialization for you. Errors like invalid keys return 401 status codes. Run out of credit triggers a 402 error.

Copyfrom openai import OpenAI

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

Streaming Responses With SSE

Stream tokens to your users for a smoother experience. Set the stream parameter to true in your request. The API returns Server-Sent Events (SSE) as the response arrives. This reduces perceived latency for long outputs.

Use the Node SDK to handle the stream efficiently. Parse the chunks as they arrive. The final chunk includes the total token usage stats. This helps you track costs accurately in real-time. Keep your connection open until the stream closes.

Copyimport OpenAI from "openai";

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

Tool Use And Function Calling

Define tools in your request to let the model call functions. The API supports standard tool definitions with parameters. Pass the function name and arguments in the response. You execute the function on your side.

Send the results back to the model in a new message. This allows for multi-turn interactions with external data. The model handles the logic for when to call tools. Use tool_choice to force or allow automatic selection. This works seamlessly with our uncensored model.

Response Format And JSON Mode

Force structured output by setting response_format to {"type": "json_object"}. The model will return valid JSON instead of plain text. This is useful for parsing data into your application. Ensure your prompt clearly describes the expected schema.

The API does not validate your JSON schema strictly. It relies on the model to produce correct syntax. If the output is malformed, you might need to retry. Use this for reliable data extraction tasks. Combine this with streaming for fast parsing.

Token Limits And Context Window

The context window supports up to 100,000 tokens total. This includes both the input prompt and the output completion. You can generate up to 16,000 tokens per request. The default max output is 2,048 tokens if not specified.

Rate limits allow 300 requests per minute per key. You can have 8 concurrent requests at once. The maximum request body size is 8 MB. Errors and refusals do not consume your prepaid credit. This makes testing safe and cost-effective.

Streaming

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)