What is an uncensored AI app?

What does 'uncensored' actually mean?

It means the model answers without refusing lawful adult, fictional, security-research, or controversial topics. Unlike standard models that often filter based on broad safety policies, this uncensored AI app focuses on raw generation capabilities. It does not mean unlimited content; there is a hard limit regarding sexual content involving minors, which is always refused. This distinction allows developers to handle their own content moderation logic if needed, or simply accept the model's broader tolerance for edge cases.

The model is open-weight and runs on our own servers. It is not GPT, Claude, Gemini, Grok, DeepSeek, Qwen, Llama, or any other vendor's model. It is a distinct entity tuned to answer without the typical refusals found in commercial APIs. You get text in, text out, with no hidden filters intercepting your prompt before the model sees it.

Why use an API instead of a chat interface?

APIs provide programmatic access, allowing you to integrate generation directly into your application's logic rather than relying on a manual web interface. This is essential for automation, batch processing, or embedding the model into larger systems like WordPress or custom dashboards. An uncensored chatbot interface is for humans; an API is for code.

Our API is OpenAI-compatible, meaning you can use the official SDKs with minimal code changes. You simply point your client to our base URL and use the provided API key. This standardization reduces integration time significantly. You are not locked into a proprietary protocol. The endpoints are straightforward: POST /v1/chat/completions for generation and GET /v1/models for metadata. No embeddings, no image generation, no fine-tuning endpoints to manage. Just raw text completion.

The apertus app model and context window

The Apertus API serves a single uncensored large language model designed for high-capacity context handling. It supports a context window of 100,000 tokens, combining both prompt and completion. This allows you to process large documents or maintain long conversations without hitting hard limits quickly. The model is tuned for lawful adult use and maintains coherence over extended interactions.

• Max output per request: 16,000 tokens (2,048 if max_tokens is not set).
• Streaming via SSE is supported, with token usage reported in the last chunk.
• Features include function calling, JSON mode, and standard parameters like temperature and top_p.

This setup is ideal for applications needing consistent, uninterrupted text generation. You do not get model routing. You get one specific, tuned model. If you need embeddings or audio, this is not the right API. It is purely for text completion.

How token pricing works

Pricing is transparent and based entirely on token usage. You pay $0.15 per 1M input tokens and $0.50 per 1M output tokens. There are no subscription tiers, no monthly fees, and no charges for idle time. Errors and refusals are free. You only pay for the tokens your prompt and completion actually consume.

Prepaid credit is charged in real-time. When your balance runs low, you top up. The credit never expires. This model eliminates the risk of paying for a subscription you do not use. It also avoids the complexity of tiered usage limits. You simply add funds and consume tokens. The rate is flat, regardless of volume or time of day.

NSFW content limits

The model is uncensored regarding adult content, fictional narratives, and controversial topics. It does not refuse these topics by default. However, there is a hard content limit that always applies: sexual content involving minors is refused. This is a non-negotiable boundary enforced by the model.

Other than this limit, the model is permissive. It will generate content that might be flagged by standard safety filters. This makes it ideal for applications where the user expects raw output. You do not need to worry about the API silently filtering your prompt. The model sees your input and generates a response based on its training, without external moderation layers interfering. This predictability is crucial for developer trust.

How to connect your app

Connecting your application is straightforward because the API is OpenAI-compatible. You need a base URL and an API key. The base URL is https://api.apertusapp.com/v1. You can use the official OpenAI SDKs or any OpenAI-compatible client by changing the base URL and providing your key.

First, sign up using Google or email/password. You receive an API key immediately. No phone number is required. Then, update your client configuration. Point it to our base URL. Send requests to /v1/chat/completions. The model ID is uncensored. That is all. You can start generating text within minutes. No complex setup or vendor onboarding is needed.

Common developer questions

Can I use this with WordPress? Yes. Any app can connect to any LLM endpoint. You just need to configure your HTTP client or SDK to point to our base URL.

What about privacy? An account needs only an email. Prompts are not used for training. Your data remains yours.

What are the limits? You get 300 requests per minute per key, 8 concurrent requests, and an 8 MB request body limit. One active key per account. If you need more, you can generate a new key, which replaces the old one.

Is there a trial? Yes. Every new account gets $0.50 of trial credit valid for 7 days. This allows you to test the API thoroughly before committing to a top-up.

Who is this for?

This API is for developers and researchers who need a single, high-capacity, unfiltered text-generation model. It is ideal for applications requiring NSFW-tolerant generation, such as adult content platforms, creative writing tools, or security research bots. If you want to avoid the noise of multiple vendor models and just need raw output, this is the right choice.

It is not for you if you need embeddings, image generation, or audio processing. It is also not for you if you require an SLA or SOC2 certification. We are an independent service. We do not claim to be an official partner of any major tech company. We provide raw token compute. If you need a chat website, we do not offer that. We provide the API layer for your own applications.

Getting started

Start by visiting the 'Get API key' page. You can sign up using Google or an email and password. Your API key is shown immediately. No phone number verification is required. You then receive $0.50 in trial credit, valid for 7 days, to test the integration.

When you are ready to scale, top up your credit using crypto. We accept USDT (TRC20) or USDC (Base). Minimum top-up is $20, maximum is $500. Bonus credits apply: +5% for $50, +10% for $100, +15% for $200. Credit never expires. You can integrate the API into your app using the standard OpenAI SDKs, pointing to our base URL. Start generating uncensored text today.