iOS app using Stable Diffusion v2 on macOS🍀
- macOS 13.1 beta 4, Xcode 14.1, MBAir/M1/8GB mem
- built for My Mac (Designed for iPad)
The app took 20 seconds for one image with 30 steps.
Try yourself. Gist.👉
gist.github.com/ynagatomo/fd87
Conversation
Replying to
The generation process of the iOS app on macOS uses 1GB memory. Interesting.🍀
1
10
Replying to
Thank you for sharing this, I have been very eager to learn about SD and try on-device. Will be trying your Gist after conversion!
1
1
Replying to
I recommend
1. create an iOS app using SD2
2. build for My Mac (Designed for iPad) and run on Mac
3. build for iPhone/iPad and run on the device lastly
Don't run on the Simulator of Xcode. It'll take a very long time.😅
1
Show replies
Replying to
If you have created your own models compatible with Stable Diffusion (ex. Dreambooth, Textual Inversion, or fine-tuning), then you need to convert the models yourself using Apple's convert tools. :)
1
1
Show replies
Replying to
The project is on GitHub.🍀
Quote Tweet
Image Generation iOS app with Stable Diffusion v2
just put the project on GitHub;
- Xcode 14.1, macOS 13.1, iOS/iPadOS 16.2
- iPhone 12+, iPad Pro M1/M2, Mac M1/M2
The project doesn't include SD2 CoreML models. See Readme.
GitHub
github.com/ynagatomo/ImgG
just put the project on GitHub;
- Xcode 14.1, macOS 13.1, iOS/iPadOS 16.2
- iPhone 12+, iPad Pro M1/M2, Mac M1/M2
The project doesn't include SD2 CoreML models. See Readme.
GitHub github.com/ynagatomo/ImgGShow this thread
1
Replying to
Could you start it on device? For some reason it produces bunch of errors and the crashes.