How do I pay for the uncensored chat API?

You top up your account using cryptocurrency, specifically USDT on TRC20 or USDC on Base. Credits are charged based on actual token usage at $0.15 per 1M input tokens and $0.50 per 1M output tokens. There are no monthly fees or subscriptions to manage.

Is this an official OpenAI service?

No, this is an independent service providing an OpenAI-compatible API. We serve a dedicated uncensored model, not GPT-4 or other vendor models. The endpoint structure matches OpenAI standards for easy integration, but the underlying infrastructure and model are distinct.

What happens if I exceed the rate limit?

You will receive a 429 status code indicating you have exceeded 300 requests per minute. You can also have up to 8 concurrent requests per key. Simply wait for the window to reset or reduce your concurrency to resume sending requests without interruption.

API Documentation - ChatUnfiltered

Authentication: Setting Your API Key

Your API key authenticates every request. You receive this key immediately after signing up with Google or an email address. Keep it secure, as it grants access to your prepaid credit. The key acts as your sole identifier for rate limits and billing. Rotate it manually via the dashboard if needed, noting that a new key replaces the old one instantly.

Base URL and Endpoint Structure

The base URL is HTTPS://api.chatunfiltered.com/v1. This structure ensures compatibility with standard OpenAI SDKs. You only need to interact with the chat completions endpoint. Embeddings, image generation, and audio processing are not available. Keep your base URL consistent across all client configurations to avoid routing errors.

curl https://api.chatunfiltered.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Check server status"}]
  }'

Making Your First POST Request

Send a POST request to /v1/chat/completions to generate text. Include your API key in the Authorization header. Set the model to "uncensored" to access the dedicated uncensored LLM. The request body must contain a messages array with your prompt. Responses return token usage and the generated content. This direct approach minimizes latency compared to multi-model proxies.

from openai import OpenAI

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

Understanding Response Formats and Streaming

Receive responses in standard JSON format or stream them via Server-Sent Events. Streaming allows you to process tokens as they generate, which improves perceived latency for long outputs. The final chunk in a stream contains the total token usage for accurate billing. You can also enforce JSON mode for structured data outputs. Choose the method that best fits your application architecture.

import OpenAI from "openai";

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

Configuring Parameters: Temperature, Top-P, and Tools

Adjust temperature and top_p to control response randomness. Use tool_choice for function calling capabilities. The model supports strict JSON mode via response_format. You can also set stop sequences to halt generation at specific points. These parameters allow fine-tuned control over the uncensored model's behavior without external routing layers.

stream = 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)

Error Handling and Token Usage Tracking

Handle 401 errors for invalid keys and 402 errors for insufficient credit. Rate limits return 429 when you exceed 300 requests per minute. Your context window spans 100,000 tokens for prompt and completion combined. Errors and refusals do not consume your prepaid credit. Track token usage in the final streaming chunk to manage costs accurately.

Uncensored LLM API Quickstart