AI Photo

AI Photo

Stable Diffusion Text-to-Image

8 ratings
20 reviews

Archived App

This is an archived listing of the app previously available on the App Store.

Although the app is no longer distributed by Apple, you can still view its description, screenshots, version history, ratings, and metadata for reference.

Price history

Details

  • Released
  • Updated
  • December 13, 2022
  • July 31, 2023
AI Photo video #1 for iPhone
AI Photo screenshot #2 for iPhone
AI Photo screenshot #3 for iPhone
AI Photo screenshot #4 for iPhone
AI Photo screenshot #5 for iPhone
AI Photo screenshot #6 for iPhone
AI Photo screenshot #7 for iPhone
iphone
ipad
🖼️Get Icon
Icons↘︎

About

Transform text descriptions into unique images and artworks offline. This app leverages advanced AI models to generate visuals directly on your device, ensuring privacy and speed. Organize, manage, and share your creations with powerful editing options.

AI-powered Text-to-Image generator
Offline image generation
Privacy-focused design
Optimized for Apple silicon
Artwork organization and management
Customizable generation parameters
Show original description

What's New in AI Photo

1.13

July 31, 2023

This update contains improvements and minor bug fixes. If you like the app, please take a moment to rate it on the App Store. Anything else you’d like to suggest? I’d love to hear from you → martin@ixeau.com

More

User reviews

VERY disappointing!

Are you excited about generative AI ? If yes, look elsewhere. Will it generate images based on your text prompts? Sure. Will you be disappointed with the result? Almost definitely. Was it worth paying for? Nope! Even if it was free I would look elsewhere. DO BETTER!

Keeps crashing

I have an IPad Pro 12. Runs CAD amazingly and fast. But this program keeps failing and crashing. Keeps telling me I’m out of memory, but I’m so far from that.

Terrible app in more ways than one

Firstly, you are only allowed to choose among already-generated AI images, which I do not care for. Secondly, it takes way too long (minute and a half or more) to generate one image, and it's not even that good. I have free apps and online generators that create AI art from any text in seconds! Not keeping this one.
Show more

Prepare to be underwhelmed!

What a disappointment! Save your money. The results are very very mediocre. I read an article that suggested it was a great apple app to leverage the power of AI art. What a joke! Sadly I'd already paid before seeing some spectacular professional looking artwork that a friend had produced. He'd used Bing which leverages ChapGPT for free!
Show more

Enjoyable way to run stable diffusion locally

I have not had any issues with the app crashing, as two other reviews mentioned months ago. So far it is very stable and a well made app with a good UI.

To the developers: is there any way that the app could have a built-in up scaler? Thanks!!

Developer apps

FAQ

What does this app do?

This app is a text-to-image generator that allows you to create unique photos and artworks based on your written descriptions. It utilizes advanced AI models, specifically Stable Diffusion, to produce visuals directly on your device.

How does the offline functionality work?

The app is designed for privacy and speed by performing all computations offline, directly on your Mac, iPhone, or iPad. This means your image generation does not require an internet connection and your data remains on your device.

What are the system requirements?

For Mac, an Apple M1 or M2 chip with at least 16 GB of RAM is recommended; Intel-based Macs are not supported. For iPhone and iPad, a current generation device (2021 or newer) with at least 8 GB of RAM is advised.

Can I use custom AI models?

On the Mac version of the app, you have the capability to add custom CoreML models. This allows for greater flexibility and personalization in the types of images you can generate.

What are the limitations of the built-in model?

The app ships with a built-in Stable Diffusion model intended for research purposes. It understands English prompts only and has limitations and biases due to its training data. Safety mechanisms are in place to limit harmful content generation.