Uncensored API
Hosted abliterated model API for developers
A single, hosted API endpoint for abliterated models that removes refusal patterns without the overhead of managing GPUs. Send prompts and receive text with full OpenAI compatibility.
- OpenAI-compatible chat endpoint
- No subscription, pay-per-token
- 100k context window support
Try it in one request
curl https://api.abliteratedmodelhub.com/v1/chat/completions \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "uncensored",
"messages": [{"role": "user", "content": "Write a blunt product review of a cheap VPN."}]
}'from openai import OpenAI
client = OpenAI(base_url="https://api.abliteratedmodelhub.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.abliteratedmodelhub.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);- $0.15
- per 1M input tokens
- $0.50
- Output tokens / 1M
- 100,000
- token context
- $0.50
- trial credit
- 300
- requests per minute
What you get
Standard Compatibility
Use the official OpenAI SDKs or any compatible client. Just update the base URL and API key to route requests to our uncensored endpoint.
Streaming & Tools
Support for Server-Sent Events (SSE) streaming and tool/function calling is built-in. Get real-time outputs and integrate complex agent workflows easily.
Transparent Pricing
Pay only for what you use. Input tokens cost $0.15 per 1M, output tokens $0.50 per 1M. Prepaid credit never expires and requires no monthly fee.
Immediate Access
Sign up with just an email and password. Your API key is displayed immediately after registration, allowing you to start testing within seconds.
Generous Limits
Each key supports 300 requests per minute and handles request bodies up to 8 MB. Regenerate your key anytime to revoke access instantly.
Privacy Focused
Prompts are not used to train the model. We keep your data usage separate from your model training pipeline for cleaner experimental results.
How it works
Create Account
Register with an email and password to receive your unique API key immediately.
Configure SDK
Point your OpenAI-compatible client to https://api.abliteratedmodelhub.com/v1 and insert your key.
Send Requests
Call the /v1/chat/completions endpoint with model ID 'uncensored' and start generating text.
What people build with it
Creative Writing
Generate unrestricted fiction, roleplay, or mature themes without the model apologizing or refusing based on content filters. Ideal for authors testing tone and voice.
Code Analysis
Query the uncensored ai model for coding tasks without refusals on controversial tech topics or aggressive debugging scenarios. Perfect for automated code review pipelines.
Security Research
Prompt injection testing and jailbreak validation become reliable when the base model does not artificially limit its own responses. Evaluate model robustness accurately.
Data Extraction
Parse large documents or logs where standard models might refuse to output certain keywords. Extract raw text data with minimal filtering interference.
What Are Abliterated Models?
Abliterated models are open-weight language models that have been processed to remove the specific layers responsible for refusal patterns. Unlike standard commercial models that are tuned to be polite and cautious, abliterated weights allow the model to answer controversial, adult, or unconventional topics without saying 'I can't do that.' This is achieved by mathematically editing the model's weights to suppress the 'refusal' behavior while retaining its core intelligence and reasoning capabilities.
These abliterated models are not merely 'uncensored' in a binary sense; they are tuned for adult use and security research where honest, unfiltered outputs are required. We host these specific weights on our own GPU servers, providing a single, consistent endpoint. You get the full benefit of an uncensored llm without needing to understand the underlying model architecture or manage GPU inference costs.
Why Use a Hosted Uncensored API?
Running an uncensored local llm requires significant hardware resources, specifically large VRAM GPUs. By using our hosted API, you offload the compute heavy-lifting to our infrastructure. You get a standard OpenAI-compatible interface, meaning you can use the same code you already know, just pointing to a different base URL.
Our endpoint supports streaming via SSE, allowing you to receive tokens in real-time. This is critical for building responsive chat interfaces or agent workflows. Unlike marketplaces that aggregate dozens of models with varying quality, we focus exclusively on one purpose-built uncensored model. This ensures consistent latency and behavior, which is vital for production applications where predictability matters.
Technical Specs: Context & Compatibility
The API supports a context window of 100,000 tokens for both prompt and completion. This allows you to process large documents, long codebases, or extended conversation histories in a single request. The endpoint strictly follows the OpenAI Chat Completions format, ensuring broad compatibility with existing tooling.
from openai import OpenAI
client = OpenAI(base_url="https://api.abliteratedmodelhub.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)Requests are limited to 8 MB in body size, and each API key can handle 300 requests per minute. If you need to rotate keys for security or load balancing, you can regenerate your key at any time, which instantly revokes the previous one. The model ID is simply 'uncensored'. There are no embeddings, image generation, or fine-tuning endpoints—just pure text completion.
Pricing: Transparent Pay-As-You-Go
We do not charge monthly fees or require subscriptions. You purchase prepaid credit, which never expires. The pricing is straightforward: $0.15 per 1 million input tokens and $0.50 per 1 million output tokens. You can top up by crypto (USDT on TRC20, USDC on Base), starting from just $20.
For larger top-ups, we offer bonus credits: +5% when you add $50 or more, and +10% when you add $100 or more, and +15% from $200. This pay-as-you-go model is ideal for developers who want to control costs precisely without committing to a recurring bill. New accounts receive $0.50 in trial credit, valid for 7 days, with no credit card required. Claim this credit with one click on your API key page.
Who This Is NOT For
This API is designed for developers who understand how LLMs work and want an uncensored endpoint. It is not for end-users looking for a ready-made chat UI or a general-purpose assistant like ChatGPT. We do not offer model routing, so you are locked into our single uncensored model. If you need embeddings, image generation, or specific SLA guarantees, this is not the right product. Additionally, while we are uncensored, we do enforce a hard limit: content involving minors is blocked. If your use case requires absolute, unfiltered content in all categories including that specific edge case, you may need a different solution.
Questions and answers
Is the model really uncensored?
Yes, the model has been abliterated to remove refusal patterns. It will answer adult, controversial, or unusual topics without polite hedging. However, we do enforce a hard content limit: requests containing sexual content involving minors are blocked. This is the only restriction that always applies.
Do I need a credit card to start?
No. You can sign up with just an email and password. New accounts receive $0.50 in trial credit that lasts for 7 days. Claim this credit with one click on your API key page. You do not need to provide a card to access this trial credit or your API key.
How does the pricing work?
We use a pay-as-you-go model with no monthly fees. You buy prepaid credit, and tokens are deducted as you use the API. Input tokens cost $0.15 per 1M, and output tokens cost $0.50 per 1M. Credit never expires.
Can I use this with the OpenAI Python SDK?
Yes. The API is fully OpenAI-compatible. You simply set the base_url to https://api.abliteratedmodelhub.com/v1 and provide your API key. The model ID to use is 'uncensored'. Streaming and tool calling are supported.
What is the context window size?
The API supports a 100,000 token context window. This includes both the input prompt and the output completion. This allows for processing large documents or long conversation histories in a single request.
Your key is one form away
Create an account, copy the key, change the base URL. That is the whole setup.
Guides
Practical articles on setup, pricing and integration.
Abliterated Model Free: Local, Demos and Trial CreditWays to try abliterated models without paying: local weights, demos and trial credit. What each costs in time and hardware, and what $0.50 of…
Abliterated Models on Hugging Face vs Hosted AccessFinding abliterated models on Hugging Face: what to check on a model card, what it takes to run the weights, and when a hosted OpenAI-compatible…
Best Abliterated Model for Coding: What to ExpectLooking for the best abliterated model for coding? See what abliteration changes, where refusals hit developers and how to test an uncensored API…
Abliterated Models on OpenRouter vs a Dedicated APIComparing abliterated models on OpenRouter with a dedicated uncensored API: model choice, consistency, setup, pricing approach and when each one…
AI Model With No Restrictions: Understanding Uncensored and Abliterated LLMsLearn what an AI model with no restrictions is. Explore abliterated models, uncensored LLMs, and how to use them for coding, creative writing, and