I’ve made a better AI video covering most of the same stuff but focusing on only one program.
“Become an AI Art legend with your iPhone, iPad or Mac ”
There are a lot of scammy AI art generators in the Mac App Store . In the video, we will cover how you can run stable diffusion on your own computer for free using three image generation apps for Mac and some of the really cool things to do.
Update: the DiffBee app has improved quite a bit from when I demoed it in this vid, to the point of being faster than the other two and more options. Be sure to click around in it as it’s the nicest UI.
My Patreon:
https://www.patreon.com/dmug
The video is a tutorial for AI Image generating on your Mac covering all the basics and even some advanced usage.
These apps are surprisingly easy to use and offer a friendly GUI. They also utilize hardware acceleration, including the GPU and, on Apple Silicon, the neural engine. With practice, you can achieve decent results.
0:00 Intro
0:38 Tip #1: Have expensive computer
1:00 App: Diffusers
1:13 AI Models
1:41 Guidance Scale
2:15 Step Count
2:57 Prompting for our first Image and refining our prompt
4:28 App: DiffusionBee
5:06 Prompt Ideas
6:17 Upscaling Images
6:48 Image to Image
7:45 Inpainting
8:36 Out Painting
10:02 App: Draw Things
10:20 Over view of Draw Things options
11:50 Stable Diffusion Samplers
12:43 Weird Results and refining
13:10 Inpainting and Outpainting in Draw Things
14:16 Making new images from existing images
A word on the implications of AI:
I have a lot of thoughts about Stable Diffusion and AI models as they’re trained often without consent on copyrighted works. This video I didn’t feel like addressing my concerns about the potential for stealing from artists. The latest version of Stable Diffusion 2.1 as I understand it reduces the reliance on stolen art. That seems like a step in the right direction.
Software Links:
DiffUsers
https://apps.apple.com/app/diffusers/id1666309574
Diffusion Bee:
https://diffusionbee.com/download
Draw Things
https://drawthings.ai/
Articles:
https://getimg.ai/guides/interactive-guide-to-stable-diffusion-guidance-scale-parameter
https://getimg.ai/guides/interactive-guide-to-stable-diffusion-steps-parameter
https://stable-diffusion-art.com/samplers/
https://www.assemblyai.com/blog/stable-diffusion-1-vs-2-what-you-need-to-know/
I've made a better AI video covering most of the same stuff but focusing on only one program. Check it out:
"Become an AI Art legend with your iPhone, iPad or Mac "
https://www.youtube.com/watch?v=cyFEvGJ0Khg
Can you please do a video on Roop also?
m1 max 32gb? mind to share the time required to process those images?
Thank you. Very interesting😊😊
Awesome thank you. I ordered a Mac Studio and assumed it would be a long time until I could do AI generated images like I can do on my Nvidia PC. Neat ! 😁👍
DrawThings look like an apple version of automatic1111.
As soon as I downloaded the app diffusers and tried to generate an image I got hit with the error
Error Domain=com.apple.CoreML Code=0 "Error computing NN outputs." Userinfo=(NSLocalizedDescription=Error computing NN outputs.}
If you could help me out on how to resolve this I would appreciate it
Will this work for a M2 base model Mac?
Great video man thanks. You should definitely do one on image and video upscaling if you get the chance.
where do you find HQ version for Diffusion Bee. On there website there is nothing.
I made another about AI that might be of interest about Chat GPT and using 3rd party clients.
https://www.youtube.com/watch?v=XGc3OHDqr5c
Holy crap! An official diffusers app????? That’s insane!
Thanks so much, just what I wanted. Very nice.
I think I like Diffusers the best, though the waifu model in DrawThings is very enticing
Great tutorial!
Amazing!
Love it!
Absolutely love that you go in-depth with your videos, this is a rare thing on YouTube. Thank you!
Thank you for sharing all of this with us! I appreciate all that you do for the community.