Quickstart: Integrate uncensored completions

First Request

Send your first completion immediately after generating your key. Use the model ID uncensored to access the open-weight model. It handles text in and text out efficiently.

The following example demonstrates a basic text completion request. Replace YOUR_API_KEY with your actual key. This confirms your authentication works correctly.

You can adjust parameters like temperature or top_p to control creativity. The model responds to standard JSON payloads. Errors return clear codes. A 402 error means you need to top up your credit. Top-ups use USDT on TRC20 or USDC on Base.

This simplicity reduces integration time. You do not need complex middleware. The API behaves like the standard OpenAI endpoints. This familiarity speeds up your development cycle significantly.

Python SDK Integration

Use the official OpenAI Python library to integrate quickly. Set the base URL and API key in the client initialization. This approach keeps your code clean and maintainable.

Pass the model name as uncensored. The library handles JSON serialization automatically. You receive structured responses that are easy to parse in your application logic.

This method works for both synchronous and asynchronous workflows. You can stream responses or get full text at once. The choice depends on your user experience needs. Streaming reduces perceived latency for long answers.

Keep your API key secure. Do not hardcode it in version control. Use environment variables for production deployments. This practice prevents accidental exposure of your credentials.

Node SDK Integration

Integrate using the Node.js SDK for server-side applications. Initialize the client with your base URL and key. The structure mirrors the Python implementation closely.

Call chat.completions.create with the uncensored model. The SDK manages the HTTP requests efficiently. You get responses in standard JSON format.

This setup works well with Express or Fastify servers. You can wrap the API call in a route handler. Handle errors gracefully to inform your users. A 429 error indicates you hit the rate limit.

Memory usage stays low because we handle the heavy lifting. Your server just processes the JSON input. This architecture scales easily with your user base. No complex infrastructure management required.

Streaming and Token Usage

Enable streaming by setting stream: true in your request. The API returns chunks of text as they are generated. This improves user experience for long responses.

Each chunk includes token usage data. The final chunk contains the total token count for the request. You can track costs in real time. This transparency helps you manage your prepaid credit effectively.

Errors during streaming are handled via standard HTTP codes. A 402 error stops the stream if credit runs out. You can resume or retry after topping up. The process is straightforward.

Streaming reduces the time to first byte. Users see progress immediately. This feature is essential for chat interfaces. It makes the uncensored LLM feel responsive and dynamic.

Rate Limits and Context Window

Our limits are generous for most use cases. You get 300 requests per minute per key. You can have 8 concurrent requests at the same time. This supports high-throughput applications without bottlenecking.

The context window is 100,000 tokens total. This includes both prompt and completion. Max output per request is 16,000 tokens. Set max_tokens explicitly to avoid defaults. The default max is 2,048 tokens.

Error codes help you debug quickly. 401 means invalid key. 402 means no credit. 429 means rate limit exceeded. Check the response body for details. Refunds are not available for credit, but mistakes are fixed via support.

Request bodies must be under 8 MB. This constraint ensures stable performance. Plan your payload size accordingly. Large documents may need chunking. This approach keeps latency low and consistent.

cURL

Copycurl https://api.openaitokenpricing.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "calculate cost for 100k tokens"}]
  }'

Python

Copyfrom openai import OpenAI

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

Node.js

Copyimport OpenAI from "openai";

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

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)