Make NSFW images in the OpenUncensored desktop app: just ask the assistant

The OpenUncensored desktop app is an uncensored AI assistant that runs on your Mac. It can read and write files, run commands, and it can draw: every one of the 17 uncensored image models on OpenUncensored is built in as a tool. You do not need to learn prompt syntax or model names. Describe what you want, and the assistant writes the prompt, picks a suitable model, generates the image and shows it in the chat.

Set it up

1. Download the app from openuncensored.com/download (macOS on Apple Silicon).
2. Buy a prepaid key at openuncensored.com/credits. No account, no email needed.
3. Open the app, paste your ou-... key, and pick a folder for the conversation. Images are saved there.

The assistant runs on an uncensored chat model (GLM 5.3 Flash Uncensored by default, and you can switch to any text model on OpenUncensored). It does not lecture you or refuse adult requests, so it writes the image prompt you asked for instead of a watered down one.

Ask for a picture

Just say what you want:

Make a photo of a woman in her late twenties in a dimly lit jazz bar, red lipstick, smoky, shot on 35mm film.

The assistant calls its built in generate_image tool. While it runs you see "Generating image", and then the picture appears in the chat. Click it to open it full size in Preview. The file is saved in the conversation's folder with a name taken from the prompt.

You can steer it like you would steer a person:

• Pick a model: "Use lustify-v8 for this" or "make it anime style with wai-illustrious".
• Change the shape: "Make it portrait" (1024x1536) or "widescreen" (1792x1024).
• More than one: "Give me four variations." Each image is billed, so the assistant makes one unless you ask for more.
• Iterate: "Same scene, but she is laughing and the lighting is warmer." The assistant rewrites the prompt and draws a new image. It is a new generation, not an edit of the old pixels, so details can change.
• Keep things out: "No tattoos, no text in the image." On the Civitai checkpoints this becomes a negative prompt.

Things the API cannot do on its own

Because the image tool lives inside an assistant that can work with files, you can combine steps in one request:

• "Write a character sheet for a fantasy succubus, then draw her portrait and a full body shot." You get the text in a markdown file and the two images next to it.
• "Read story.md and illustrate each of the five scenes." The assistant reads your file, writes one prompt per scene and saves scene-1.png to scene-5.png.
• "Make a 3x3 grid of the same outfit in nine colours." It generates the images and, if you ask, stitches them into one file with a command line tool.
• "Rename everything in this folder by the model used and move the anime ones into anime/."

What it costs

The app is free. You only pay for what you generate: each image is billed from your credits at the model's price, typically one to five cents (see the live prices on the models page), plus the text tokens the assistant uses. Your prepaid balance is a hard cap, so you never get a surprise bill.

Prefer code? Use the API

The same models, the same key, one OpenAI compatible call:

curl https://openuncensored.com/api/v1/images/generations \
  -H "Authorization: Bearer $OU_API_KEY" -H "Content-Type: application/json" \
  -d '{"model": "lustify-v8", "prompt": "boudoir photo, soft window light, 85mm, film grain", "size": "1024x1536"}'

The NSFW image API guide covers the parameters, the Python SDK and which model to use.

What is allowed

Adult content between adults is allowed on every model listed here. Three things are not, and the providers refuse them: anything involving minors or anyone who looks underage, real people (celebrities, people you know, lookalikes) without their consent, and non consensual imagery. If you publish what you make, label it as AI generated and follow the rules of the site you post on.