Uncensored API
Abliterated llm models for unrestricted output
Get the abliterated LLM models that scrub refusal directions from weights for unrestricted output up to 100,000 tokens per request.
curl https://api.abliteratedllmmodels.com/v1/chat/completions \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "uncensored",
"messages": [{"role": "user", "content": "Show the status of the deployment pipeline"}]
}'from openai import OpenAI
client = OpenAI(base_url="https://api.abliteratedllmmodels.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)import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://api.abliteratedllmmodels.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);OpenAI-compatible chat completions
Streaming SSE support
Crypto top-ups only
Key features
Unrestricted Output
Get uncensored responses for lawful adult topics without standard refusals.
Large Context Window
Process 100,000 tokens including prompt and completion together.
Streaming Support
Receive tokens via SSE with usage data in the final chunk.
Function Calling
Use tools and tool_choice with standard JSON mode support.
No Subscription
Pay only for real token usage with prepaid credit that never expires.
Crypto Payments
Top up with USDT or USDC and receive bonus credits instantly.
Setup guide
Get API Key
Sign up with Google or email to receive your key immediately.
Add Crypto Credit
Top up via USDT or USDC to start using the API.
Send Requests
Use the uncensored model ID with your OpenAI compatible client.
Real-world use cases
Content Creation
Generate unrestricted NSFW uncensored stories without hitting standard filters.
AI Research
Explore model biases by comparing abliterated models against standard versions.
Developer Testing
Test function calling and JSON mode with an open-weight model.
Raw Data Extraction
Extract text from large documents using the 100k token context.
abliterated LLM models for unrestricted content generation
We deliver abliterated LLM models that remove refusal directions from weights, yielding uncensored AI models without losing intelligence. Use this NSFW uncensored endpoint for creative writing, adult fiction, or raw data extraction.
Adult Fiction & Roleplay
Generate unrestricted narratives without polite hedging or sudden topic shifts.
Security Research
Extract raw outputs from controversial prompts without content filters blocking your queries.
Creative Writing
Output unfiltered prose for novels or scripts that need mature themes.
Data Extraction
Pull text from diverse sources without artificial summarization or safety truncation.
From signup to first token in under two minutes
Start receiving unrestricted outputs immediately without waiting for approval.
- Register. Click Get API key and sign in with Google or an email and password. The system displays your key instantly.
- Load credit. Top up via USDT (TRC20) or USDC (Base).
Switch your OpenAI base_url in three lines of code
Change your base URL and API key to start sending requests immediately. Our API is fully compatible with standard OpenAI SDKs, so you can swap providers without rewriting your application logic. Simply point your client to HTTPS://api.abliteratedllmmodels.com/v1 and use the "uncensored" model ID. The transition is fast, keeping your existing code structure intact while removing content filters.
This approach works for any OpenAI-compatible client.
Why most "uncensored" models still filter you — and how to avoid it
They filter because they are merely uncensored, not abliterated. Standard models retain refusal weights, causing unpredictable blocks during complex prompts. Abliterated models mathematically scrub the direction vector from the weights, ensuring unrestricted outputs without degrading base intelligence. Avoid proxies that route traffic; we run one dedicated model on our own servers. Use the correct model ID to guarantee consistent, unfiltered behavior.
Who this API is for (and who should stick to GPT)
Developers needing unrestricted text generation use this API for creative writing, roleplay, and security research where standard filters block nuanced or adult topics. Our abliterated model preserves intelligence without refusing lawful content.
- Best for: NSFW uncensored stories, unfiltered chatbot logic, and raw data extraction.
- Not for: Teams needing image generation, embeddings, or strict corporate compliance. If you require GPT’s specific ecosystem features, stick to GPT.
Exact cost math: how much your prompts will really cost
Input costs $0.15 per 1M tokens and output is $0.50 per 1M tokens. A 10k prompt plus 1k output uses 0.011M tokens total, totaling $0.00165. Errors and refusals are free. No monthly fee. Paid credit never expires. Use our calculator to verify your exact spend before topping up with USDT or USDC.
Common questions
What exactly are abliterated LLM models?
They are models with guardrails mathematically scrubbed from their weights. This process removes refusal direction so you get unrestricted outputs without losing the base intelligence of the underlying architecture or degrading performance.
Does this API work with standard OpenAI SDKs?
Yes. It is fully OpenAI-compatible. You simply update your base URL to our endpoint and provide your API key. This works with the official SDKs and any standard OpenAI-compatible client you already use.
What is the context window limit?
The model supports a 100,000 token context window for both prompt and completion. You can send up to 16,000 tokens per request, or 2,048 tokens if you do not explicitly set the max_tokens parameter.
How do I pay for the API?
You top up with crypto only, specifically USDT on TRC20 or USDC on Base. Any whole amount from $20 to $500 is accepted.
Which payment methods do you accept?
Crypto only: USDT on TRC20 or USDC on Base, any whole amount from $20 to $500. Bonus credit applies from $50.
Can I use this for function calling?
Yes. The API supports tools, tool_choice, and JSON mode via the response_format parameter. You can also use streaming via SSE, set temperature, top_p, stop sequences, and adjust presence and frequency penalties effectively.
Get your API key now
Sign up, copy the token, update your base URL. That is it.
