Ablated API Documentation

Base URL & Authentication

Start every request against https://api.ablated.cc/v1 using your unique key. We provide one active key per account, issued instantly after you sign up with Google or email. There is no phone number required, no subscription tier, and no monthly fee to worry about. Your credit is prepaid and never expires, so you only pay for what you actually use. Errors and refusals consume zero tokens, meaning you are not charged for mistakes. This model is NOT GPT, Claude, or any other vendor's model; it is a distinct open-weight model tuned for uncensored output.

First Request

Send a standard chat completion to get your first uncensored response. The endpoint accepts JSON payloads with your model id and prompt. You do not need to specify complex routing or multiple models. The system routes directly to our single ablated model. This simplicity ensures consistent behavior without hidden tiers. Use the following curl command to test connectivity immediately.

Copycurl https://api.ablated.cc/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?"}]
  }'

Python SDK

Integrate the ablated LLM into your existing Python workflows with minimal code. The official OpenAI SDK works perfectly because we maintain the same interface. You only need to update the base URL and provide your API key. This approach lets you keep your existing logic while gaining access to an uncensored model. The SDK handles serialization and retries automatically. Configure the client once, then call the completions endpoint as usual.

Copyfrom openai import OpenAI

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

Deploy the ablated model in your JavaScript or TypeScript projects effortlessly. The Node SDK mirrors the Python experience, requiring only a base URL change. This compatibility allows you to swap in the uncensored AI without rewriting your entire application logic. You get raw text output, ideal for creative or NSFW use cases. The library handles the HTTP requests and response parsing for you.

Copyimport OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.ablated.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 & Token Usage

Receive tokens in real-time via Server-Sent Events (SSE) for lower latency. The final chunk of the stream contains the total token usage for billing purposes. This feature is essential for applications that need to display text as it generates. You can also enable JSON mode by setting the response format to json_object. This ensures the output is valid JSON, which is useful for structured data extraction. Streaming works exactly like the standard OpenAI streaming protocol.

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)

Limits & Parameters

Each key supports 300 requests per minute and 8 concurrent requests. The context window is 100,000 tokens for prompt and completion combined. If you do not set max_tokens, the default output is 2,048 tokens. You can adjust temperature, top_p, and other standard parameters to control creativity. A 401 error means your key is invalid, while 402 indicates insufficient credit. The 429 rate limit appears if you exceed the per-minute threshold. The hard content limit always applies: sexual content involving minors is refused.