How to Set up Stable Diffusion AI on Mac

How to Set up Stable Diffusion AI on Mac - AUTOMATIC1111

Рет қаралды 36,458

Күн бұрын

I will walk you through the step-by-step process of setting up Stable Diffusion Web UI locally on your Mac M1 or M2.
🔗 Installation guide: github.com/TechXplainator/Tut...
🔗 Dreamshaper Model: civitai.com/models/4384/dream...
___________________________________________________________________________________________________
Udemy Courses:
🎓 Stable Diffusion: www.udemy.com/course/your-ult...
🎓 Leonardo AI: www.udemy.com/course/your-ult...
🎓 Prompt Engineering: www.udemy.com/course/prompt-e...
🎓 Blockchain: www.udemy.com/course/your-ult...
___________________________________________________________________________________________________
KZfaq Playlists:
📺 Leonardo AI: • Leonardo - the AI imag...
📺 Prompt Engineering: • Prompt Engineering
📺 Stable Diffusion: • Stable Diffusion
📺 Mac Tips: • 🍏 Mac Tips
📺 Canva: • Canva
📺 AI News: • AI News
📺 AI - The Big Reset: • AI - The big Reset
📺 AI Text-to-Video Generation Tool Comparison: • AI Text-to-Video Gener...
___________________________________________________________________________________________________
Connect with Me:
🌐 Website: www.techxplainator.com/
👋🏻 Instagram: / techxplainator
👋🏻 TikTok: / techxplainator
___________________________________________________________________________________________________
0:00 Intro
0:12 Installing Stable Diffusion
4:50 Running Stable Diffusion
5:58 Ending Stable Diffusion
6:21 Conclusion
___________________________________________________________________________________________________
#stablediffusion #aiart #howto #tutorial

Пікірлер: 167

@SERAPIS50 3 ай бұрын

I'm just repeating what everyone else has already stated. Superb tutorial! I've wanted to download stable diffusion for ages but it just looked too complicated. You're a treasure, keep up the good work.

@FThighlights2023 5 ай бұрын

You broke it down nicely. Really helpful. The best tutorial I've seen so far. Thanks

@abysm4010 8 ай бұрын

Thanks for the clear tutorial! It's been really helpful

@Eleiem 7 ай бұрын

This is an amazing tutorial. Thought had everything installed yet was having troubles launching it, the solution was the PATH command I missed. Tyvm

@Boxdue 9 ай бұрын

Fantastic! finally SD in the Mac world. I hope you continue🤩

@TechXplainator 9 ай бұрын

Thanks so much for taking the time to comment! And yes, I will continue :-)

@Piczoid 6 ай бұрын

Great tutorial, worked perfectly on MacOS Ventura

@m76r72 6 ай бұрын

Great tutorial, thanks!

@crd2204 4 ай бұрын

Thank you so much, very good tutorial

@khalilnascimento 7 ай бұрын

You are just amazing! Thanks a lot!

@reboreji9312 3 ай бұрын

Excellent tutorial.....briliant work

@niKolasBedoya 8 ай бұрын

Great video, thanks

@karstenkoch4655 6 ай бұрын

Thanks. This is great tutorial. Its short, its great, its working. Happy New Year

@TechXplainator 6 ай бұрын

You're welcome! A happy new year to you as well!

@karstenkoch4655 6 ай бұрын

👍

@Diegouzzz 4 ай бұрын

Thank you. the only tutorial that worked for me!

@TechXplainator 4 ай бұрын

Glad it helped!

@eduhorta 6 ай бұрын

Great video! Everything working as expected!! Btw.. Juggernaut XL is a great model to try!

@TechXplainator 6 ай бұрын

Thanks, I really appreciate it. I never used Juggernaut XL before, but I definitely will give it a try. Thanks for sharing!

@wengxiaoxiong666 4 ай бұрын

thanks , it's very useful!!!

@dudewithpictures 4 ай бұрын

Best tutorial ever!

@b1gdan 5 ай бұрын

very helpful thanks alot

@cogidigm 8 ай бұрын

Thank you for thsi video.

@Liminally_ill 4 ай бұрын

This tutorial is so user friendly and also NOT ten million years long! I had to comment and like, such an awesome video

@TechXplainator 4 ай бұрын

Wow, thanks!

@jessesnowden 6 ай бұрын

thank you!!!

@indianshek 8 ай бұрын

Huge Thanks for your clear video. may i ask do i need to download SDXL separately id i wanted to do mostly in XL??? Cheerssss

@TechXplainator 8 ай бұрын

You're very welcome. Yes, you need some more steps for SDXL. There are some dependencies must be downloaded together with the model file. Honestly: so far, I have not been able to use SDXL on mac without issues, especially since I have limited RAM (8GB). But recently, there were some SDXL models that have been published which are optimized for Ma. So I will try to use those instead. As soon as I get them to work, I will post a video.

@Bboreal88 4 ай бұрын

Hey! Such a nice explanation and super glad someone is teaching AI on Mac. But I have a question, do you have any video teaching how to fine-tune SD on Mac? I want to create a model with my own data, how do I do that?

@TechXplainator 4 ай бұрын

Hey! Thanks for the kind words! I don't have a video on fine-tuning SD yet, but I'm open to suggestions! I'll definitely put that on the list of future videos. If you have any other questions or ideas, feel free to share!

@user-xi4li7in4o 7 ай бұрын

Thanks for the great video, but I've got the error NansException: A tensor with all NaNs was produced in Unet. This could be either because there's not enough precision to represent the picture, or because your video card does not support half type. Try setting the "Upcast cross attention layer to float32" option in Settings > Stable Diffusion or using the --no-half commandline argument to fix this. Use --disable-nan-check commandline argument to disable this check. I've tried to use "--no-half" and "--disable-nan-check", but not working, I've also tried to turn off the MPS "export PYTORCH_ENABLE_MPS_FALLBACK=1", failed to create the image. I'm using M2 ultra, does anyone know how to fix this？ Thank you!!

@TechXplainator 7 ай бұрын

Thanks for reaching out! I appreciate your feedback. This might fix your issue: github.com/AUTOMATIC1111/stable-diffusion-webui/discussions/5461#discussioncomment-7533743. I hope it helps!

@RJMCTV 8 ай бұрын

Thanks for the video. Another question. To install additional models. Can I add them to the models folder whilst the Terminal and/or browser UI is still running, or should I quit out of it, add the models to the folder and then restart

@TechXplainator 8 ай бұрын

Thanks! That's a great question. You should definitely quit the UI, add the models to the folder and then restart via terminal app. Otherwise the UI will not recognize the new files.

@RJMCTV 8 ай бұрын

Thanks :) @@TechXplainator

@TechXplainator 8 ай бұрын

I have to correct myself. The refresh-button beside the model drop down works just fine (it didn't previously). This means: There is no need to quit SD after installing a new model. Refreshing is sufficient.

@tiagol924 4 ай бұрын

funciona mesmo!

@content1 5 ай бұрын

IN CASE IT HELPS to anyone, when you type the first line in the terminal, do wait because it seems is not working but it takes a while to ask you for the passport

@quizkraftstudio 2 ай бұрын

I just get this error meassge in the terminal when I try and actually create an image "No module 'xformers'. Proceeding without it. Warning: caught exception 'Torch not compiled with CUDA enabled', memory monitor disabled" any ideas?

@TechXplainator 2 ай бұрын

XFormers exclusively functions with Nvidia cards. If you're using a Mac, experiment with the "--opt-sub-quad-attention" option to generate larger images. For further information, visit this link: github.com/AUTOMATIC1111/stable-diffusion-webui/discussions/7453

@aryanhawoldar8348 6 ай бұрын

Hey, great tutorial but at the end when I type cd stable-diffusion-webui into terminal it says "no such file" even though the folder is in my downloads. What should I do

@TechXplainator 6 ай бұрын

Hi there! I'm sorry to hear about the issue you're facing. I've created a troubleshooting guide specifically for the 'no such file' error. Check it out here: techxplainator.github.io/stable-diffusion/troubleshooting.html#2-error-no-such-file-when-typing-cd-stable-diffusion-webui I hope it helps you resolve the problem. Best of luck, and I hope it works smoothly for you!"

@THiran-ep2vj 7 ай бұрын

Hi! I'm getting the issue TypeError: Cannot convert a MPS Tensor to float64 dtype as the MPS framework doesn't support float64. Please use float32 instead. Is there any fix?

@TechXplainator 7 ай бұрын

I'm not familiar with this error. Does it happen with all models? Or only with a specific checkpoint file?

@virasatsingh3014 27 күн бұрын

Great video! Any clue why I might be getting the following error? 'NoneType' object has no attribute 'lowvram'

@TechXplainator 27 күн бұрын

Thanks, I appreciate it! Regarding the error: I found this bug report on the AUTOMATIC1111 Github page for someone who had the same issue. There were 2 suggestions, but I don't know if they worked: 1. Update your mac (github.com/AUTOMATIC1111/stable-diffusion-webui/issues/15637#issuecomment-2094152779). 2. Update command line arguments (github.com/AUTOMATIC1111/stable-diffusion-webui/issues/15637#issuecomment-2095943078) - also see this thread for more details: github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Command-Line-Arguments-and-Settings. I hope this helps ☺️

@Creaturessa 8 ай бұрын

Thank you, this was hugely helpful. I previously tried to instal from some other instructions and did an Unattended installation of Homebrew. Then got stuck and started all over again with your instructions. While opening SD I got the following warning: Warning: caught exception 'Torch not compiled with CUDA enabled', memory monitor disabled What does it mean? Help please :) Thank you.

@TechXplainator 8 ай бұрын

Thank you for taking the time to comment, and I'm glad my video helped. The warning "caught exception 'Torch not compiled with CUDA enabled', memory monitor disabled" only means that SD Web UI didn't detect any NVIDIA card, which is expected on mac. For clarification: Stable Diffusion Web UI works best with NVIDIA graphics card, but the developers at AUTOMATIC1111 added some features that make it run efficiently on Mac's M1/M2 as well. Still, this warning comes up. But it shouldn't be an issue since it is only a warning, not an error message. Stable Diffusion Web UI should run fine regardless. I hope this helps :-)

@Creaturessa 8 ай бұрын

@@TechXplainator no, THANK YOU, for taking the time to respond in such detail. Can you please point me to good tutorials to 1- Use Stable diffusion for stills (I use Midjourney since the beginning, but never really dig into SD) and 2 - Use SD for animation. Greatly appreciated. Thank you again :)

@TechXplainator 8 ай бұрын

@@Creaturessa you're very welcome☺. For stills I can recommend my own course on Udemy, its a very basic course for beginners and I've just shared 100 free coupons on my community tab, so you can just sign up for free 🙂. Here's the link to the post: kzfaq.infoUgkxvL9sZQspyX1nUrpLlZtkGoD--eXTZByh . For animation, I don't have a good recommendation, but I will try to share a video about that here on youtube soon.

@Creaturessa 8 ай бұрын

@@TechXplainator you’re the best! Thank you 😊✨

@ahmadataei70 7 ай бұрын

thanks for the tutorial. will it work properly on my 2020 macbook pro m1, it has 8gb ram?

@TechXplainator 7 ай бұрын

It will run, however it's not optimal. I also have 8 GB of RAM and image generation gets super slow at times.

@devacore477 4 ай бұрын

Agree, you need a powerful pc to run SD on your system. Mine actually froze using it with similar specs. @@TechXplainator

@stableArtAI Ай бұрын

After a few months and many tweaks it is better to be doing local than online like leonardo. Somethings might be a little, say, mechanical. But still very cool, you guides has been so much help, thanks again. The web, does have a few limits with respect to styles. But the style editor is a big plus in organization of using them. So naming them can be important too. We don't really know the effects of running the latest torch vision with torch 2.1.2 because it list 0.16.2 to be the one, yet we are running 0.17 version. As well as we have not installed a working version of torch 2.2.x as yet. Nor have we update python version( may recall might have did. new version that automatic and SD 1.5 does not officially support to work with lora's and checkpoints.(for a latter time, still processing so many images to share). Current which seems pretty stable for us: version: v1.9.4 • python: 3.10.14 • torch: 2.1.2 • xformers: N/A • gradio: 3.41.2

@TechXplainator Ай бұрын

Thanks for the feedback! I'm glad my guides have been helpful. Running local has its advantages, especially with style organization. Your current setup looks solid! Keep up the great work and feel free to share more updates. Meanwhile, I've been focusing on local LLMs (as you may have noticed from my recent videos ☺️), but I want to dive back into SD, especially img2video, soon.

@stableArtAI Ай бұрын

@@TechXplainator image to video seem still kind of weak(that is the default extensions available, unless you know of a better one).

@stableArtAI Ай бұрын

@@TechXplainator solid, yes, seen a big difference from the original version and after tweaking all the errors from python and torch. Even dreamsamplerXL works better. still though confused on the SD 2 and 3 which seem to be all integrated into the updates however ckpts and lora would need to be compatible.

@amyzhang1701 4 ай бұрын

Thank you for the tutorial. When I opened the stable diffusion on Mac it says "TypeError: Cannot convert a MPS Tensor to float64 dtype as the MPS framework doesn't support float64. Please use float32 instead." do you know how to fix it?

@TechXplainator 4 ай бұрын

Thanks! And sorry to hear about the issues you ran into. I'm not familiar with this error myself, but I found this thread on GitHub that discusses a fix - described here: github.com/AUTOMATIC1111/stable-diffusion-webui/issues/12907#issuecomment-1873561734. I hope it helps!

@industructable7275 6 ай бұрын

Hey there, I am stuck on the URL bit, trying to download the folder because when I type in " git clone URL" it comes up with a message saying fatal: destination path 'stable-diffusion-webui' already exists and is not an empty directory. But I have checked my finder and downloads hundreds of times and I cannot find the stable diffusion webui folder. What is wrong?

@TechXplainator 6 ай бұрын

I appreciate you reaching out with the issue you're facing. It seems like you're encountering a common problem when trying to clone a Git repository. You can quickly resolve this by following the steps outlined in my troubleshooting guide: techxplainator.github.io/stable-diffusion/troubleshooting.html I hope this helps!

@FThighlights2023 5 ай бұрын

I encountered quite some issues running the codes. At some point I had to delete some directories and start all over again. This wasn't accounted for in your post. So what could be the problem?

@TechXplainator 5 ай бұрын

Sorry to hear that. It could be that Stable Diffusion has made some significant changes since the time I published this video which causes some issues now. This is an open source software that is constantly updated by the community. They have a running issue for Mac here: github.com/AUTOMATIC1111/stable-diffusion-webui/discussions/5461. Perhaps you can find your problem described there? Good luck!

@FFTuk Ай бұрын

I managed to get to 3:07 cloning the web Uk repository but I can’t find the folder in finder under home. If I try and repeat the process it says fatal: destination path ‘stable diffusion web ui’ already exists and is not an empty directory. Any ideas where it is?

@TechXplainator Ай бұрын

Normally you'll find the cloned repository under "home". If not, please have a look at this troubleshooting guide I've created which addresses the issue: techxplainator.github.io/stable-diffusion/troubleshooting.html I hope this helps ☺️

@ztp2130 4 ай бұрын

Great video. Could you address the hardware requirements to run Stable Diffusion on a Mac? For example, I have a MacBook Air M2 with 16 GB of memory. Will that be enough power to run this service, including training models?

@TechXplainator 4 ай бұрын

Thank you very much! Below are the official hardware requirements for running Stable Diffusion, including the web UI from Automatic 1111: 1. macOS 12.5.1 or later. 2. Apple Silicon Mac model (M1 M2, or M3 chip recommended). 3. Minimum 16 GB of RAM, with 8 GB as the absolute minimum. 4. Intel-based Mac CPUs with GPU acceleration (performance may be slower). From personal experience (using an M2 Mac with 8 GB RAM), I found that generating images with generation one and two models was relatively straightforward. However, I encountered significant challenges when attempting to use SDXL (for more information, refer to this video: kzfaq.info/get/bejne/iriWZ7OSma_KemQ.html). Unfortunately, local model training was not successful for me due to insufficient RAM. As a workaround, I successfully executed the model training process on Google Colab, though this option requires a paid plan. I hope that helps ☺️

@ztp2130 4 ай бұрын

@@TechXplainator Awesome! Thank you very much. You explain things very clearly. Subscribed!

@garen591 3 ай бұрын

@@TechXplainator When you say using generation one and two models, do you mean based on the 1.5 stable diffusion models? Cause i want to use Rev_Animated models which is based on 1.5 models. Generally how fast is it to generate 512x512 images and batches of 512x512 images on a mac mini M2 ?

@TechXplainator 3 ай бұрын

Yes, precisely. I'm referring to models based on SD 1.5 or SD 2.1. The generation speed varies depending on your GPU. With a MacBook Air M2 with 16GB of memory, I would estimate that generating a single 512x512 image could take around 60-90 seconds. However, I haven't personally tested this configuration, so it's based on my research. If you're generating batches, it will likely take longer. Please keep this in mind.

@frederickphang1917 9 ай бұрын

Hi, I encountered problem like "× Preparing metadata (pyproject.toml) did not run successfully." and "error: metadata-generation-failed × Encountered error while generating package metadata." Do you know how to resolve it?

@TechXplainator 9 ай бұрын

I have not encountered this issue personally, but it sounds like it can be resolved by upgrading pip (the python package installer). Please see this repair suggestion for more info - I hope it helps!: github.com/AUTOMATIC1111/stable-diffusion-webui/issues/2491#issuecomment-1314434462

@synergy9268 5 ай бұрын

Hi TechXplainator ! I have this message when trying to create an image : "NansException: A tensor with all NaNs was produced in Unet. This could be either because there's not enough precision to represent the picture, or because your video card does not support half type. Try setting the "Upcast cross attention layer to float32" option in Settings > Stable Diffusion or using the --no-half commandline argument to fix this. Use --disable-nan-check commandline argument to disable this check. Time taken: 3.7 sec." I made some suggested manipulations but it"s still buggy. Any idea ? I'm on it since a few hours, tried several time the same suggestions I find on internet... Starting to annoy me. Thank you for your work.

@TechXplainator 5 ай бұрын

Sorry to hear that. I found this potential fix online: github.com/AUTOMATIC1111/stable-diffusion-webui/discussions/5461#discussioncomment-7989071. Hope this helps!

@jameskim3232 3 ай бұрын

I cant generate anything because it says "AssertionError: Torch not compiled with CUDA enabled". From other sources, it says that i should still run, but it still isn't. Problem might be with the Torch version error? I need help.

@TechXplainator 3 ай бұрын

Thanks for reaching out. The message 'Torch not compiled with CUDA enabled' means that PyTorch isn't detecting an NVIDIA graphics card, which is common for Mac users. This message is more of a warning than an error, so you should still be able to run Stable Diffusion on your Mac. However, are you encountering any other errors? I'd recommend checking the Terminal application for additional error messages that might give us more insight.

@stfranck8691 3 ай бұрын

Hello I've got this error "Repo already cloned, using it as install directory" = ModuleNotFoundError: No module named 'importlib.metadata' this is after trying to run Stable - any idea? this is around 5 min of your video - thanks for your help

@TechXplainator 3 ай бұрын

Thanks for the info. The message "Repo already cloned, using it as install directory" isn't an error; it means the script is using an existing repository for installation. For "ModuleNotFoundError: No module named 'importlib.metadata'" on Mac, try (in the terminal app before running SD): pip install importlib-metadata Hope this helps!

@user-jb6ws5ie8m 2 ай бұрын

the url doesnt show on the terminal, is there any chance i can see it int he folder?

@TechXplainator 2 ай бұрын

The URL for the web UI remains constant at 127.0.0.1:7860. You can bookmark this link in your browser but it will only work after running ./webui.sh in your terminal.

@clementfilmmaker 5 ай бұрын

Hi, Great tutorial, at the end of your video you don't explain how to restart stable diffusion once you close it.

@TechXplainator 5 ай бұрын

Thanks! No worries. You would start stable diffusion the same way via terminal app using the command ./webui.sh

@URNNGKULL 5 ай бұрын

Thanks for the tutorial, really appreciate it. It seems that I did everything accordingly, but for some reason when I try to generate something I get two error messages: connection errored out and python quit unexpectedly. You have any idea how can I fix it?

@TechXplainator 5 ай бұрын

Unfortunately, Stable Diffusion works best with NVIDIA GPUs, so it has limitations on Macs. Here's what you can do: 1. Make your images smaller to ease the workload and avoid timeouts. 2. Try using an older model like gen 1, based on Stable Diffusion v1.5. 3. If you're still having issues after resizing and using an older model, try reinstalling Stable Diffusion. Hope this helps!

@URNNGKULL 5 ай бұрын

@@TechXplainator How do I use an older model, I did everything according to your instructions

@TechXplainator 5 ай бұрын

Oh okay. Its difficult to know how to fix these kinds of errors from afar. Sorry, I'm out of ideas then. Perhaps someone else in the KZfaq community can help? Good luck!

@Alexfernando2255 7 ай бұрын

can i install in macbook pro 2.3 GHz 8-Core Intel Core i9 | radeon pro 560X 4 GB | 16 GB Ram

@TechXplainator 7 ай бұрын

Thanks for reaching out! I haven't tried it on an Intel Core, but according to AUTOMATIC1111, It should be possible. Happy Installing 🙂

@matteobonaldo3992 5 ай бұрын

Thanks for the tutorial! I managed to install everything perfectly and start SD. But when I try to generate a picture though it says this "RuntimeError: "LayerNormKernelImpl" not implemented for 'Half'", how can I resolve this? :,(

@TechXplainator 4 ай бұрын

Sorry to hear that. I found this issue on GitHub where this issue was solved: github.com/AUTOMATIC1111/stable-diffusion-webui/discussions/10981#discussioncomment-6133633. I hope this helps!

@matteobonaldo3992 4 ай бұрын

@@TechXplainator Everything works perfectly! God bless you for your precious help

@chircami 4 ай бұрын

Amazing tutorial. Hello, I have just followed all the steps, I open the program but I get this error: AssertionError: Torch not compiled with CUDA enabled. Has anyone solved it or has the solution? I would appreciate it, best regards

@TechXplainator 4 ай бұрын

Thanks for your comment, and I'm glad you found the tutorial helpful! The message 'Torch not compiled with CUDA enabled' simply indicates that no NVIDIA card was found, which is typical for Macs. It's just a warning, not an error, so Stable Diffusion should run properly on your Mac. Hope that clears things up! 😊

@devacore477 4 ай бұрын

Thanks for your tutorial, but I wouldn't advise using this on osx Sonoma, it really messed with my system. I got it going only to crash with errors after errors. What is the best way to delete all SD and all the associate files from my system?

@TechXplainator 4 ай бұрын

To uninstall the downloaded GitHub repository, you'll need to follow these steps in reverse order of the installation process: 1. Begin by navigating to Finder and deleting the folder named "stable-diffusion-webui." This action should also remove the models stored in the "model" folder. 2. Next, uninstall all programs installed via Homebrew by executing the command: brew uninstall cmake protobuf rust python@3.10 git wget If you wish to uninstall Homebrew entirely, please refer to the guide provided here: docs.brew.sh/FAQ#how-do-i-uninstall-homebrew

@devacore477 4 ай бұрын

Thanks so much, appreciated. Could you advise on what system comp specs are best for SD 2.1? @@TechXplainator

@TechXplainator 4 ай бұрын

Actually, I was able to run SD 2.1 on Mac M1 mini with 8GB RAM. It wasn't quick - in fact it took several minutes to generate an image, but it worked. So I'm surprised at why you had so many issues. The minimum specs are: M1 with 8GB Ram for Gen 1 and Gen 2 models. But in general: more RAM would be better.

@ilsarak332 6 ай бұрын

hi sorry, it said to me " RuntimeError: Placeholder storage has not been allocated on MPS device!". How can I fix this?

@TechXplainator 6 ай бұрын

Are you using controlnet? I haven't encountered that error yet, but I found a link that describes a fix: github.com/Mikubill/sd-webui-controlnet/issues/1128. Hope this helps!

@synergy9268 5 ай бұрын

Hi @@TechXplainator I have this message when trying to create an image : "NansException: A tensor with all NaNs was produced in Unet. This could be either because there's not enough precision to represent the picture, or because your video card does not support half type. Try setting the "Upcast cross attention layer to float32" option in Settings > Stable Diffusion or using the --no-half commandline argument to fix this. Use --disable-nan-check commandline argument to disable this check. Time taken: 3.7 sec." I made some suggested manipulations but it"s still buggy. Any idea ? I'm on it since a few hours, tried several time the same suggestions I find on internet... Starting to annoy me. Thank you for your work.

@quynam9 4 ай бұрын

how to install deforum on mac

@1edber 7 ай бұрын

any luck installing Stable Video Diffusion locally on an apple silicon based Mac? thanks! :D

@TechXplainator 7 ай бұрын

Yes. In this video I'm installing it on an M2 Macobook air. I have also successfully installed it on a Mac Mini M1 using the same process I showed in the video. In the video description, you'll also find a step-by-step instruction. I hope this helps!

@1edber 7 ай бұрын

@@TechXplainator Hi there, thank you for your response. This video shows the instal for Stable Diffusion. I'm asking about the new Stable Video Diffusion that was recently released. I would love to see how the installation of it would be on a Mac. If. you ever make such a video I would race to watch it! Thanks!

@TechXplainator 7 ай бұрын

Oh, I see. No, I haven't tried that yet. But if I do, I will be sure to make a video about it. Thanks for your kind comment :-)

@1edber 7 ай бұрын

Awesome, thanks!! @@TechXplainator

@AleynaB-bp9bv 8 ай бұрын

Does this installation come with controlnet? If not, is it possible to use controlnet with macOS?

@TechXplainator 8 ай бұрын

Thanks for reaching out! No, you'll need to install ControlNet separately as an extension. And yes, I was able to use ControlNet with my M1/M2 even with low VRAM (8GB). However, it is very slow. I will soon make a video about ControlNet. Meanwhile, please find a quick installation guide on my GitHub page: techxplainator.github.io/stable-diffusion/install-controlnet.html

@AleynaB-bp9bv 8 ай бұрын

@@TechXplainator Thank you!

@plasticboi2 6 ай бұрын

Your instructions are so clear that I don’t know how I got so messed up on Step 2. I get zsh command not found: brew. I’m unable to delete or proceed without the word brew. What am I doing wrong. Thank you

@TechXplainator 6 ай бұрын

Hi there, no worries! Please follow the instructions on this page to fix the issue: mac.install.guide/homebrew/zsh-command-not-found-brew.html. Hope it work! 🍀

@plasticboi2 6 ай бұрын

@@TechXplainator Thank you for responding so quickly. I’ll try it tomorrow when I fish my laptop out of the pond 😆

@TechXplainator 6 ай бұрын

🤣

@plasticboi2 5 ай бұрын

I followed your instructions and created a stable-diffusion-webui folder. Thannk you. @@TechXplainator

@plasticboi2 5 ай бұрын

@@TechXplainator It worked! Thank you for taking the time to help! 🙏🏾✌🏾❤️✨

@MartinNiemoller-bp3yj 6 ай бұрын

Error Expecting value: line 1 column 1 (char 0). why the stable diffusion show me this

@TechXplainator 6 ай бұрын

Hi there, I found this thread that discusses this problem and this suggestion seems to fix it: github.com/AUTOMATIC1111/stable-diffusion-webui/issues/9150#issuecomment-1489775463. Hope this helps!

@beautyofchaos585 4 ай бұрын

Ther's an error dquote cmdsubst>. Anybody to help me, please?

@TechXplainator 4 ай бұрын

Hey! So, it seems this might have occurred while installing Homebrew, correct? It probably happened because the path variable wasn't copied correctly. To resolve this, take a look at this documentation that explains how to properly set up your path variable: sourabhbajaj.com/mac-setup/Homebrew/. Best of luck!

@akierskan7787 Ай бұрын

Hi! Could you please do a tutorial on this Stable Diffusion for image 2 video or text 2 video?

@TechXplainator Ай бұрын

Sure, I’d be happy to! Stay tuned for a tutorial on Stable Diffusion for image-to-video and text-to-video. Thanks for the suggestion!

@akierskan7787 Ай бұрын

@@TechXplainator thank you so much! ❤️❤️

@akierskan7787 Ай бұрын

@@TechXplainator Hi, sorry, it's me again. :( I tried generating images but this issue seems to come up: AttributeError: 'NoneType' object has no attribute 'lowvram'? Would you have any advice to fix this? :( I really wanna start generating images. :(

@TechXplainator Ай бұрын

I found this bug report on the AUTOMATIC1111 Github page for someone who had the same issue. There were 2 suggestions, but I don't know if they worked: 1. Update your mac (github.com/AUTOMATIC1111/stable-diffusion-webui/issues/15637#issuecomment-2094152779). 2. Update command line arguments (github.com/AUTOMATIC1111/stable-diffusion-webui/issues/15637#issuecomment-2095943078) - also see this thread for more details: github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Command-Line-Arguments-and-Settings. I hope this helps ☺️

@akierskan7787 Ай бұрын

@@TechXplainator oh my gosh! You’re amazing! Thank you so much! 😭😭😭 Appreciate you being at the forefront of Stable Diffusion for Macs!

@alanatyson3633 2 ай бұрын

why does it keep saying “attribute error: ‘nonetype’ object has no attribute’lowvram’

@TechXplainator 2 ай бұрын

When does that happen? When generating an image?

@nikolaysparkov Ай бұрын

I got an error - cd: string not in pwd: run

@TechXplainator Ай бұрын

when are you getting this error / at which step? cd stands for "choose directory" and this error often occurs when you have a typo in the directory you are trying to open. Please check out this page for a good description: www.howtouselinux.com/post/fix-cd-string-not-in-pwd. I hope this helps!

@Kunalalal 4 ай бұрын

will it cause damage to my Mac in anyway?

@TechXplainator 4 ай бұрын

Great question, thank you! Automatic1111 is an open-source software that undergoes continuous review by numerous developers worldwide, significantly reducing the likelihood of it containing harmful bugs or viruses. However, like any tool downloaded from the internet, there's no absolute guarantee of safety. The same applies to prerequisite tools like Python, Homebrew, and others.

@devacore477 4 ай бұрын

it broke my system - wouldn't advise on an M1 Mac.

@TechXplainator 4 ай бұрын

Sorry to hear that. It was not my experience (Mac M1 mini - 8GB RAM). I'm working on a video on alternative ways to run Stable Diffusion on Mac, specifically for low GPU macs. Stable Diffusion Web UI is more suitable for very high powered Macs and better alternatives have become available.

@mr.secretd1682 26 күн бұрын

how could i remove this program from my mac ? it take to much place

@TechXplainator 26 күн бұрын

@TheMagista88 7 ай бұрын

No m3?

@TechXplainator 7 ай бұрын

I assume it works for M3 as well, although I haven't tried.

@AndrewDavidBaron 20 күн бұрын

I spent entire day trying this. Kept getting error messages. Tried to uninstall and reinstall every aspect of this tutorial. I even uninstalled and reinstalled Terminal itself. Nothing worked. I gave up. Literally wasted an entire day.

@KiiDbeamO 5 ай бұрын

My Mac is not the best but I create Photos of buildings and liminal stuff… any chance you’d know or anyone If this can run on my Mac?… my processor: 2GHz Quad-Core Intel Core i5. Graphics: Intel Iris Plus Graphics 1536MB. Memory: 16GB 3722MHz LPDDR4X MacOS Sonoma 14.2.1

@TechXplainator 5 ай бұрын

hmmm... not sure. I have only tried it with M1/M2 processors. Sorry.

@stableArtAI 5 ай бұрын

basically it should run on any, there I a way to user cpu memory, SD is designed for GPU but can run on CPU too. As a neophyte the is one of the best install guide found and worked like charm. Since started our journey we made some big strides.

@Jakobono22 7 ай бұрын

why can't i enter my password for the installation in terminal? :D

@TechXplainator 7 ай бұрын

When you're in the terminal app and its asking for password, you just start typing and then hit enter. You will not see a window or a field for the password, just type ahead and enter, then the installation process continues. Hope that helps ☺️

@Jakobono22 7 ай бұрын

Thx a lot :)@@TechXplainator

@Jakobono22 7 ай бұрын

So everything went fineuntil this error during opening SD: ImportError: dlopen(/Users/jaquesack/stable-diffusion-webui/venv/lib/python3.9/site-packages/cv2/cv2.abi3.so, 2): Symbol not found: ___darwin_check_fd_set_overflow Referenced from: /Users/jaquesack/stable-diffusion-webui/venv/lib/python3.9/site-packages/cv2/.dylibs/libX11.6.dylib (which was built for Mac OS X 11.0) Expected in: /usr/lib/libSystem.B.dylib in /Users/jaquesack/stable-diffusion-webui/venv/lib/python3.9/site-packages/cv2/.dylibs/libX11.6.dylib Could you maybe help me in any way? Thx in advance and for the great content

@TechXplainator 7 ай бұрын

Thanks for your great comment 💙. It looks like you are on python version 3.9 - perhaps from an earlier installation? Please make sure to install at least version 3.10 - or higher. You may use the command behind this installation guide to get python 3.10: techxplainator.github.io/stable-diffusion/sd-installation-guide.html#step-2---use-homebrew-to-install-required-applications. Good luck 🍀

@Moe-zl1ph 6 ай бұрын

in a few months it´s gonna be just one click, and done.

@gamerfitness7895 6 ай бұрын

Doesn’t work on Mac air

@TechXplainator 6 ай бұрын

Hi there, actually: it works for me on a MacBook Air M2 2022. What's the issue you're facing?

@gamerfitness7895 6 ай бұрын

I have the M1 chip and I keep getting error for multiple processes after adding the model. I can’t get the URL every single time that I tried 🤷🏽‍♂️

@TechXplainator 6 ай бұрын

Sorry to hear that. If it helps, here's a list of common issues and solutions with Mac installation, please feel free to check if your errors are among them: github.com/AUTOMATIC1111/stable-diffusion-webui/discussions/5461 Good luck!

@Junjunterkombek 4 ай бұрын

Test

@lfkessler 8 ай бұрын

Hi! thanks for the tutorial but I received a message when I tryed to generate a image RuntimeError: "upsample_nearest2d_channels_last" not implemented for 'Half'

@TechXplainator 8 ай бұрын

Thank! I didn't get the error myself and I was not able to reproduce it. But I found this thread on Github with a potential solution: github.com/AUTOMATIC1111/stable-diffusion-webui/issues/8555#issuecomment-1606270779. I hope this helps!

@lfkessler 8 ай бұрын

@@TechXplainator WOW! thanks a Lot! It's working now!!!! Thanks! Thanks!

@user-ip1um7fu2o 4 ай бұрын

Hi after I copy and paste this line nothing is happening (echo; echo 'eval "$(/usr/local/bin/brew shellenv)"') >> /Users/BG/.zprofile eval "$(/usr/local/bin/brew shellenv)". Does it take a while do I just need to let it sit and wait? What do I look for so I know it's done and I can start Step 2 the Install Applications. Also when I hit enter the first time I never saw a Warning message like you showed. I did see ==> Installation successful! but if I scroll down a little I do see this ==> Next steps: - Run these two commands in your terminal to add Homebrew to your PATH: (echo; echo 'eval "$(/usr/local/bin/brew shellenv)"') >> /Users/BG/.zprofile eval "$(/usr/local/bin/brew shellenv)" But again nothing happens.

@TechXplainator 4 ай бұрын

Certainly! After pasting the command into the terminal, it should execute immediately. You'll know it's done when you see the cursor again, usually within a few seconds. To confirm Homebrew is installed, type `brew -v` in the terminal. If it shows the version, Homebrew is installed. If not, consider downloading it from Homebrew's GitHub page: [Homebrew Releases](github.com/Homebrew/brew/releases/tag/4.2.8). Then proceed with step 2. I hope that helps 🙂

@nicolasmejia1736 7 ай бұрын

everything was going ok I was getting this until the error message .... Resolving deltas: 100% (152/152), done. Installing requirements for CodeFormer Installing requirements Launching Web UI with arguments: --skip-torch-cuda-test --upcast-sampling --no-half-vae --use-cpu interrogate no module 'xformers'. Processing without... no module 'xformers'. Processing without... Traceback (most recent call last): File "launch.py", line 48, in main() File "launch.py", line 44, in main start() File "/Users/nmg/stable-diffusion-webui/modules/launch_utils.py", line 432, in start import webui File "/Users/nmg/stable-diffusion-webui/webui.py", line 13, in initialize.imports() File "/Users/nmg/stable-diffusion-webui/modules/initialize.py", line 33, in imports from modules import shared_init File "/Users/nmg/stable-diffusion-webui/modules/shared_init.py", line 5, in from modules import shared File "/Users/nmg/stable-diffusion-webui/modules/shared.py", line 5, in from modules import shared_cmd_options, shared_gradio_themes, options, shared_items, sd_models_types File "/Users/nmg/stable-diffusion-webui/modules/options.py", line 74, in class Options: File "/Users/nmg/stable-diffusion-webui/modules/options.py", line 77, in Options def __init__(self, data_labels: dict[str, OptionInfo], restricted_opts): TypeError: 'type' object is not subscriptable

@TechXplainator 7 ай бұрын

Could be a conflict with a potential previous Python version you had installed, but I'm not sure. Can you please make sure you are using at least python version 3.10 for stable diffusion? To check the version, go to terminal and type in: python3 --version. It should give you the version. If you still get the error in spite of having version 3.10 (or higher), please open an issue in the AUTOMATIC1111 GitHub repo: github.com/AUTOMATIC1111/stable-diffusion-webui/issues. I've looked for your error message and couldn't find a good solution, so this may be worthy of a new issue. Good luck!