hi every time I write "brew install cmake protobuf rust python@3.10 git wget" on the terminal it says "command not found" I'm blocked I can not install stable diffusion.
@LaCarnevali Жыл бұрын
Hey, I assume you installed Homebrew. If so, it can be that Homebrew was installed in another directory - you can check that when you run the Homebrew installation on your terminal. In my case, Homebrew was installed into /opt/homebrew: " ==> This script will install: /opt/homebrew/bin/brew /opt/homebrew/share/doc/homebrew /opt/homebrew/share/man/man1/brew.1 /opt/homebrew/share/zsh/site-functions/_brew " If you get another directory, you have to run brew from its origin. For macOS Intel, the directory is /usr/local, so in that case, you should run the below: /usr/local/bin/brew install cmake protobuf rust python@3.10 git wget Then follow NEXT STEPS shown on command line, it should show something like this ==> Next steps: - Add Homebrew to your PATH in /Users/$USER/.zprofile: echo 'eval $(/opt/homebrew/bin/brew shellenv)' >> /Users/$USER/.zprofile eval $(/opt/homebrew/bin/brew shellenv) Copy and paste the first line on the same terminal window, then hit enter echo 'eval $(/opt/homebrew/bin/brew shellenv)' >> /Users/$USER/.zprofile Then Copy and paste the last line on the same terminal window, then hit enter eval $(/opt/homebrew/bin/brew shellenv) Finally you should be able to use brew help and if that works, it means that brew was installed properly and running from ZSH.
@soumayaseghir4177 Жыл бұрын
@@LaCarnevali hi thank you very much for your answer. I followed your instruction and your video for installing stable diffusion and it worked. it took little bit time but it's done. 🥰
@LaCarnevali Жыл бұрын
@@soumayaseghir4177 Happy it worked!! 😄🥰
@WildWorldTravelers Жыл бұрын
@@LaCarnevali can you help me?? How you did it??
@enocorxl Жыл бұрын
i have the same problem
@bimyou2794 Жыл бұрын
After hours and hours .... also several times trying.. trial and error , 😭😭😭finally it works because of you!!thank you Laura 🔥🔥🔥
@Ying2022lan8 ай бұрын
Hello thank you very much for the tutorial. but every time I am installing it I run into this error" TypeError: unsupported operand type(s) for |: 'type' and 'NoneType'. may I know how to resolve it?
@2ndEarth Жыл бұрын
I looked in many different places to find good download instructions, your video is by far the best one, great job!
@snaessens Жыл бұрын
Thank you Laura for this tutorial. I couldn't have installed Stable Diffusion on my Mac without your video! Just subscribed to your channel!
@ellegolding793810 ай бұрын
Managed to get it done after a few errors but got there at the end. Thank you for your video. Very helpful.
@LYREMusicGroup Жыл бұрын
Thank you so, so much for this tutorial! Terminal usually terrifies me, but I was able to follow the steps and install it like a pro, thanks to you!
@ShadowbannedMedia Жыл бұрын
How do you reopen it after you save it to the browser? Thanks in advance
@esmeraldosinforme Жыл бұрын
Hey Laura Carnevali! I am from Brazil. Just wanted to say a big thank you for all your help. Your channel is awesome, and I've been telling my friends about it. Keep up the great work!
@jejakpotret Жыл бұрын
Hey, it's really helpfull and works perfect as well, I've been tried many tutorials on mac always failed and complicated, your tutorial are simple, clear, easy and amazing, thanks a lot
@nadiatremoulet2977 Жыл бұрын
After several hours and several attempts BOOM...mil GRACIAS. Without you I would not have succeeded ☺
@AppaloosaMichael Жыл бұрын
Hi Laura, thanks for the video. Have you considered making a video on how to create an alias application to run stable diffusion? that would allow users to click an icon from the dock instead of manually navigating to the terminal and running the .sh command
@VoxerArt Жыл бұрын
could you guide me how to do this?
@allensong6693 Жыл бұрын
thank you so much. Laura, i finally installed sd on my mac. it's working perfectly. many thanks.Best to you.
@izcarus Жыл бұрын
Laura! You are amazing, I subscribed as thanks for taking the time to post and respond. EDIT UPDATE: I was getting an error message that "ERROR: Could not find a version that satisfies the requirement torch==1.12.1" ---- Turns out I uploaded Python 3.11 which is not supported and when I uploaded 3.10 it seems to be working. Still looking forward to more tutorials if you get into detail about different model packages. THX!
@LaCarnevali Жыл бұрын
Thank you, Carlos! I m going to post soon a video to fix these small warnings when running SD! Happy you sorted :)
@nicolas.c Жыл бұрын
worked like a charm! best explanation straight to the point, thanks!👏
@UnrealOcean Жыл бұрын
thank u laura, your instruction to use Warp to install made this very easy!
@aaron7006 Жыл бұрын
Though there were some hiccups during the installation process due to network connectivity issues, I managed to complete it successfully by following the video step by step. It's fantastic! Thank you!
@filmreifberaten11 ай бұрын
You are an angel. Worked like a charm. Thank you so much from Switzerland
@Ruslan54RUSАй бұрын
Thank you so much, I've installed SD and ran it. It was hard but successfully haha
@nikolaiskiba2294 Жыл бұрын
Hi Laura! Thank you for instruction! just installed on my MacBook Pro 14 m1 Pro according to your instructions. everything is working! there was just a warning "Torch not compiled with CUDA enabled". and there was also a notice "A new release of pop is available: 23.0.1 - 23.1.2" to update run: pip instal --upgrade pip (and after installation I typed this but nothing happened.. maybe it has updated already? thank you for your work Laura!
@LaCarnevali Жыл бұрын
Hi, these are just warnings and are not relevant for Mac users. For more info >>> kzbin.info/www/bejne/q5aQkHWKZaqnn7s
@TONG34567 Жыл бұрын
woo. your article help me a lot! I have searched KZbin for week. Thanks a lot!
@JanAnton Жыл бұрын
You have just gained a new follower, Laura 😊 thanks for a very nice tutorial
@subtledagger3052 Жыл бұрын
Worked perfectly. Thank you very much!
@mikopetras8607 Жыл бұрын
Thank you, Laura, it worked perfectly. I needed to go for details and links to your page :)
@rigorojas Жыл бұрын
Thank you, your clear instructions worked for me.
@oliversimon9800 Жыл бұрын
i don't have a mac, and i have stable-diffusion for now installed on a debian PC, i watch the video only because i like you :)
@LaCarnevali Жыл бұрын
lol
@warpdotdev Жыл бұрын
Great video. Let us know if you have any feedback or questions while using the Warp terminal!
@atony229 Жыл бұрын
Worked exactly as you instructed. Thank you!
@EdvvinWang Жыл бұрын
So so thankful for this video installation tutorials. Appreciate. Warp terminal works!! Everything else works, except my stable diffusion doesn't generate any image hahaha...
@LaCarnevali Жыл бұрын
Lol, Hi! What error is showing on the terminal when you press the generate button?
@EdvvinWang Жыл бұрын
So thankful dear. If it wasn't you. None of this would happen.
@justsomegirlwithoutamustache5 ай бұрын
same issue here lol
@bardoof6 ай бұрын
Thank you so much. It works without a problem.
@harmonicdissent Жыл бұрын
Thank you for making this. You approach was very detailed and simple. It's rare that things like this are so easy. You have a wonderful day.
@arson36311 ай бұрын
Big Thanks!I can really install stable diffusion on mac! Yayyyy
@kikilei2605 Жыл бұрын
Wow thank you very much for sharing the amazing tutorial! It worked perfectly😊
@Adrian_Galilea Жыл бұрын
Hey Laura, thanks for the tuto, I was wondering what was the performance on the m1/m2
@LaCarnevali Жыл бұрын
It can be very slow. Although there are some steps you could take to improve the performance when starting SD. I would recommend to use google colab
@CelestiC Жыл бұрын
Thanks a lot great tutorial ! Time to unleash my creativity :D
@milenakuzior3083 Жыл бұрын
Laura, you help me so much! thank you !
@TheFoodyFrog Жыл бұрын
Thank you so much Laura I spent so much time but it works 🙏
@ismo3974 Жыл бұрын
Amazing video, it worked perfectly on my end as well
@CaviarTaco Жыл бұрын
Awesome tutorial! Thank u Laura 🤗
@fred4838 Жыл бұрын
Thank you Luara just installed. How can I link my output folder to export to my external drive?
@oliwan-kenobi3 ай бұрын
Hi Laura, thanks a lot for the nice video :) Is stable-diffusion-v1-5 still the latest and best version to install?
@LaCarnevali3 ай бұрын
It depends on what you want to achieve. If you need to use ControNet, it probably is :)
@carolcross_ArtFusionStudio Жыл бұрын
Thank you. This is what I was looking for. I want to create some of the liquid, dreamy, animations.
@its.harlow Жыл бұрын
Hi, thanks for sharing. I've however ran into an error stating that "ERROR: Could not find a version that satisfies the requirement torch==1.12.1 (from versions: none) ERROR: No matching distribution found for torch==1.12.1" how to I trouble shoot this?
@LaCarnevali Жыл бұрын
Hi Joanna, you need to upgrade torch (probably need change environment). You can watch this video at min 6:30 as I am looking into this error and solving it: kzbin.info/www/bejne/q5aQkHWKZaqnn7s
@aligirl42043 ай бұрын
Is this video outdated and the command brew install cmake protobuf rust python@3.10 git wget doesn't work anymore? Because every time I do copy paste, is says "command not found". What should I do? Thank you
@SD_Alias Жыл бұрын
Hi as a mid 50s retoucher that started retouching in the 80s with brushes and airbrushes this is very scary to see. I have witnessed the development of my profession over the last 30 years with a mixture of horror and fascination. And I'm afraid that is another development that can't be stopped. So I can't avoid dealing with it in my old age. Can stable diffusion be used without programming knowledge? I don't know where to start. 😔 I am a Photoshop user on a Mac Studio but i have no IT skills. Or would it be better to start with other software ? I just can't spend a lot of money unfortunately.... Are there any beginner tutorials?
@LaCarnevali Жыл бұрын
Hi, I see what you mean. It can be scary. The artist has to be still there, taking their creative thoughts and translating them onto the "page"; it’s just that the tools accommodating that process are different. And you're going in the right direction trying this. I think using Midjourney or Dall-e 2 is easier, but they are not 100% free. Stable Diffusion is 100% free. You should probably start installing it on Google Colab - the installation process is way easier, and the image generation speed is way faster. Regarding tutorials, I would look into KZbin tutorials. Once you play a bit with it, it's not difficult. Google Colab tutorial: kzbin.info/www/bejne/q3exq2evbaqVsJY
@SD_Alias Жыл бұрын
@@LaCarnevali Thank you very much! It is the first time i hear about google Colab. Is this something like Google Docs? I will try it next days as you suggest and have a look at this tutorial...
@fullflowstudios Жыл бұрын
If you are old what am I then? I'm 70 and heavily involved in Stable Diffusion and in getting the CodeFormer AI routines to work locally because of the AMAZING facial restoration capabilities. I suspect that it will lead me to a great online business in building graphics with real people from the past in A.I. environments that never got to be( like Berlin without the second World War). If you were thinking that the last three years of covid were a big disruption in your fifties than buckle up for the next three years!! A.I. will be a lot more disruptive than covid for sure. Just read this comment, three years from now and check if I'm right. Laura is looking at it the right way: she's such a gem. Cristal clear guidelines along a strong and erudite train of thoughts while still being so creative. I really admire her. She is an example of a human being that knows: "And at once I knew I am not magnificent"(according to the lyrics of Holocene by Bon Iver).
@RiverReeves23 Жыл бұрын
Great video! Can you please make a video on how to train a LORA locally on a Mac M1? This seems to have close to no documentation and people are struggling. I found one guy on Reddit that got it running and can train in 10 mins but everyone else is running in to brick walls.
@LaCarnevali Жыл бұрын
Cannot train a model on mac m1 to my knowledge :)
@ddude2 Жыл бұрын
Thanks for a very clear and simple video Laura. Is the Core ML version o stable diffusion faster than the PyTorch version. I am using Apple MacBook Pro with M2 chip. If so, how do I also install CoreML version of SD 1.5? Thank You
@LaCarnevali Жыл бұрын
Hi, same here as per Mochi. I need to test them before being able to say something about them. They can be faster, but not sure in terms of flexibility?
@olivermensa8106 Жыл бұрын
Hey laura! everything went quite smooth but now we're trying to generate our first image and..... RuntimeError: "LayerNormKernelImpl" not implemented for 'Half' any ideas?😬
@LaCarnevali Жыл бұрын
Hi Oliver, In either the webui-user.bat or webui-user.sh add the below to the COMMANDLINE_ARGS= variable: --precision full --no-half This should work. Let me know :)
@Sniper18211 Жыл бұрын
@@LaCarnevali had the same problem and tried this fix but unfortunately didn't work :(
@LaCarnevali Жыл бұрын
@@Sniper18211 Have you updated to latest macOS?
@Sniper18211 Жыл бұрын
@@LaCarnevali tbh no, still on monterey 12.2.1 :/
@Sniper18211 Жыл бұрын
sounds like we need to restart from 0? 🤣
@LeWorm45 Жыл бұрын
Hi, I just finished watching your video about LoRAs and it was really helpful, and same with this one! I am wondering if you could help me, what specs does the mac have that you use for Stable Diffusion? Would a MacBook Pro M2 13" do the job?
@LaCarnevali Жыл бұрын
I m not using my mac as mine (M1) is not powerful enough. M2 should be fine though!
@trant01019 ай бұрын
Hi i'm unable to generate images. Getting this pop up" 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'm using Mac mini M1
@3ig3rand Жыл бұрын
Muchas gracias Laura, me ayudaste un montón!!!
@BigChangeGmbH Жыл бұрын
this was so helpful thanks so much!!!
@MiguelFolgado-u3t Жыл бұрын
Very cool! Thanks. Do you know if it is possible to train a Lora using this app on Mac?
@LaCarnevali Жыл бұрын
Hi Miguel, training a model requires lot of power, so probably best using Colab or similar.
@tommyLee-rb1bb7 ай бұрын
hey Laura, thanks for the vid, very clear and helpful. i followed ur steps and ran the webui.sh in terminal and it opens a webpage with the ui automatically however within the terminal it says the stable diffusion failed to load and using the ui i am unable to generate an image where its say AttributeError: 'NoneType' object has no attribute 'pop'. Could u please help with this, thanks.
@LaCarnevali7 ай бұрын
Try to run the requirements file before the webui.sh. pip install -r requirements.txt
@tommyLee-rb1bb7 ай бұрын
@@LaCarnevali Thank you for replying, i think the problem was that i hadnt installed the brew correctly and it woudlnt let me unless i shut the terminal completely first after downloading homebrew and then i could install it. But it works like a charm, i appreciate ur help!
@SN-em9hx5 ай бұрын
Thank you! 2024 and it still works!
@melissavanderree3655 Жыл бұрын
Thank you so much! Most effective tutorial ever on yt! :)
@michaelcottam9519 Жыл бұрын
This did not work out as expected. But it has been a big help! I'm having issues with git saying files can't be checked out and many other problems. Now I'm stuck with the Clip file not loading. Looks like there is something not quite right at Git. It takes a lot longer to download all these files too, about three hours and the video is only 7:18 just so you know. But thanks for the video, I would have never known how to even try this.
@LaCarnevali Жыл бұрын
Hi Micheal, very weird. You might want to try Colab? See link below. kzbin.info/www/bejne/q3exq2evbaqVsJY
@Glazedribeye Жыл бұрын
Hi, great tutorial. One thing I'd like to know, is M1 capable of running controlnet? Do you intend to make a video about it?
@LaCarnevali Жыл бұрын
Hi, the performance on M1 is not great, I would recommend to use an external gpu like colab/runpod
@lalpremi Жыл бұрын
Thank you, have a great day :-)
@Blaco-l2c Жыл бұрын
It works perfect, thank you so much!
@Tyrell_ai Жыл бұрын
thank you so much! it works perfectly even on m2 Max apple
@merlinvongarnier1802 Жыл бұрын
Thank you so much for the video
@kantaklaro44735 ай бұрын
Hi Laura, I have quick question. On a i Mac M3 should I follow he same procedure described on this vid? Thanks!
@ChocoSunsei Жыл бұрын
Thank you ! Your video is very helpful :D
@jdotf8999 Жыл бұрын
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: 37.6 sec.
@LaCarnevali Жыл бұрын
try running the webui with the --no-half argument: ./webui.sh --no-half
@justsomegirlwithoutamustache5 ай бұрын
I followed all the instructions step by step and everything went perfect with the installation process. Then it came to actually generating images where I was met with the message"AttributeError: 'NoneType' object has no attribute 'lowvram'". I don't know how to fix it or why it's happening
@kasetia21 Жыл бұрын
Hi Laura. Many thanks for sharing. How can I know my github version and how to update it on Mac M1?
@vincentdoms2712 Жыл бұрын
Very very nice and helpful ! Thx 😊
@THEARPE07 Жыл бұрын
great vid thank you very much for taking time and your effort, not many smart and pretty ladies in tech business :) thanks
@rawstersnazer7645 Жыл бұрын
Hi laura, Thanks for the instructions. Does the same process work for Intel Macs? I have an Intel i9 processor with 32GB ram & AMD Radeon Pro 5500M 4 GB graphic card.
@ShyBoyEnt Жыл бұрын
How well will this run on a base model M2 Mac silicon?
@LaCarnevali Жыл бұрын
I dont have an M2 to test it, lol, but people with the M2 seems to be okay with that.
@jasonyang8007 Жыл бұрын
Thank you so much, so clear and easy to follow :)
@John_Krone Жыл бұрын
I heard Windows users enter something into the Webui batch that allows to upgrade automatically in Automatic "eleven eleven", can we do the same with mac?? Thank you for this tutorial. subscribed!!
@LaCarnevali Жыл бұрын
I m scared that on Mac you need to git pull if you want to update it. I will double check :)
@doloreslagrange Жыл бұрын
hello, great video! i have istalled it and it worked like a charmed! but i have close it , and now i don´t know how to reopen it again! help and thanks!
@LaCarnevali Жыл бұрын
open the terminal, cd into the main folder, ./webui.sh
@doloreslagrange Жыл бұрын
trank you so much!!!!@@LaCarnevali
@cjsim2 Жыл бұрын
You are a gem.
@JimmyHoussen Жыл бұрын
Thank you for your video, it allowed me to install Stab-eDiffusion on my M1 Macbook Air. I have also installed Deforum + Automatic111 and have been using it for about 1 week. The 3D option does not work for me, starts to render and quits. Did you happen to solve this issue? Seems to be an issue with memory and I probably will need a more powerful computer. Thanks again.
@PopoRamos Жыл бұрын
great tutorial! have you got the training to work? i have been trying to train the AI on my face but it keeps crashing when i start the training process.
@LaCarnevali Жыл бұрын
I tried on Colab and worked okay. Traning the model requires lot of power.
@PopoRamos Жыл бұрын
@@LaCarnevali i'm using an 1M ultra 128gb ram. But I just went ahead and used my windows machine with a 3090ti in it. It blows the mac out of the water in terms of GPU speed.
@fullflowstudios Жыл бұрын
CUDA's is what you need Popo. Just go with the Nvidia's !
@PopoRamos Жыл бұрын
@@fullflowstudios Yeah i figured that out real quick! i been using my windows machine for all AI stuff now, its so much faster!
@ProductionsIndeed Жыл бұрын
hello! thanks for the info, this is some of the best out there right now. Just a heads up, when I try to copy code from your Warp console its hard to see what the commands are. This only comes up when I'm paused tho. Cheers!
@LaCarnevali Жыл бұрын
Thank you, Jeremiah! Noted :)
@oliverzhang6102 Жыл бұрын
Thanks! That's really interesting!
@joseph-rennie9 ай бұрын
Hi Laura, im getting this error after I hit on generate in the SD UI 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 did - Upcast cross attention layer to float32 but still receiving this error. Can you help me on this?
@numeirr8 ай бұрын
Thank you so muchhhhhhhh i subbed and liked
@alxsiberian Жыл бұрын
Super easy and useful. Thanks
@Rabhishekphotography Жыл бұрын
Hi Laura Love from India.Thankyou for making for such info. videos!! i just wanted to ask if this works only on M1 Mac's or works on Intel mac too. asking before trying!!
@LaCarnevali Жыл бұрын
Hi!!! Yes it does! The only difference is that Brew is going to be installed in another directory wrt to the video, so you need to use some additional code - but still easy! If you want to give a try, let me know if you need any help :)
@kyocatwoman Жыл бұрын
how to i turn it on again if i close the terminal at some point?
@LaCarnevali Жыл бұрын
open the terminal, go into the main folder using the cd command, and type: ./webui.sh
@syruptear7620 Жыл бұрын
Great Video! Thank you so much!
@Beetoe Жыл бұрын
Grazie mille Laura!! Tutorial perfetto!
@raaaaah-gh2pi Жыл бұрын
Thank you, very helpful.
@007deetee Жыл бұрын
Excellent tutorial thank you :)
@TheSoggetto Жыл бұрын
Ciao laura! Grazie per il video, super utile. Una domanda: in caso volessi istallare midjourney? Come lo installo?
@LaCarnevali Жыл бұрын
midjourney non va installato, vai su Discord e attivalo :)
@TheSoggetto Жыл бұрын
@@LaCarnevali Grazie! Non ho capito: posso attivare Midjourney e utilizzarlo in locale attraverso il terminale? Inoltre per ogni esportazione (attraverso Stable Diffusion) ci metto anche piu di due minuti se 512x512. Credo ci sia un problema forse?
@Buinhikl7 ай бұрын
hi from germany, thank you so much for the tutorial. I got it installed and running just fine, but the next day, the web doesnt load. I copy paste the link and refresh, but it says Safari cant connect to the server. What should i do?
@EdmnDnts Жыл бұрын
Thanks, what about starting it from terminal? i've installed but how to start it the next day the same process?
@LaCarnevali Жыл бұрын
need to only run the webui.sh file
@jamesbriggs Жыл бұрын
great video!
@Fiveash-Art8 ай бұрын
This install looks like a major pain.. I was looking at some version of stable diffusion for Mac 'Draw Things' which looks like a regular download. Is that comparable to whatever version you're using in the video?
@yalulin Жыл бұрын
thank you for this very helpful tutorial. i do wonder how i can set up with cloud GPU such as lambda, just for better flexibility with extensions and i cannot seem to find tutorial online
@LaCarnevali Жыл бұрын
You can use Google Colab (colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast_stable_diffusion_AUTOMATIC1111.ipynb) I ll make a video next week ;)
@Korfmaker Жыл бұрын
Question, what's the first line of code I need to type in the Terminal, after everything worked fine? Because now I all the time need to redo the whole installation process, while most files are already in place of course. I guess my question is very similar to the one from Laura Carnevali, but I didn't manage so far...
@LaCarnevali Жыл бұрын
Hi, you don't need to rerun the all file, unless you want to update the folder from gitHub. To run Stable Diffusion you need to go into the main directory: >>> cd desktop/stable-diffusion/stable-diffusion-webui (cd ) then run the ./webui.sh file >>> ./webui.sh
@Korfmaker Жыл бұрын
@@LaCarnevali That worked, thank you very much Laura! Being a graphic designer, the Terminal and Warp are a bit of unknown territory for me. But thanks to you I managed :)
@korakistudios Жыл бұрын
hello, it did worked , but once the terminal is closed, where and how do I open stable diffusion after that?please help
@LaCarnevali Жыл бұрын
need to run the ./webui.sh file using the terminal
@bimyou2794 Жыл бұрын
hey laura i try to run local on imac, .. got this error msg .. "Cannot convert a MPS Tensor to float64 dtype as the MPS framework doesn't support float64. Please use float32 instead." can you help me ?
@BigRedJphoto Жыл бұрын
really good, but one question, i closed terminal, and then it stops, oops, what do i do to run agian? and not install? thanks for the great video
@LaCarnevali Жыл бұрын
need to open the terminal again, dive in the main folder (cd stable-diffusion-webui) and run the webui file (./webui.sh)
@ArtByOmarNYC Жыл бұрын
thanks for making this video! unfortunately im getting this error... can you please help?..... RuntimeError: "LayerNormKernelImpl" not implemented for 'Half'
@LaCarnevali Жыл бұрын
Hi Omar, Have you updated to the latest MacOS? If yes, try the following: In either the webui-user.bat or webui-user.sh add the below to the COMMANDLINE_ARGS= variable: --precision full --no-half Let me know if it works :)
@dmitrykarpov7936 ай бұрын
What is the best settings to improve velocity of work for Mac M1?
@LaCarnevali4 ай бұрын
not to use the mac M1 eheh. Sorry, I would love to give a different answer, but SD on mac at the moment is a big NO. Too heavy. You should go with Colab/RunPod or DiffusionHub/ThinkDiffusion etc.
@ORGANICMINDART Жыл бұрын
hi .. Thank U for the tutorial mac stable diffusion .. could U also make tutorial for training LORA in mac M1 please ?😇
@LaCarnevali Жыл бұрын
Hi! You can use this video: kzbin.info/www/bejne/rpnSnYVsZ7Zlqrs However, I doubt you might be able to run a training on your M1 as not powerful enough (I have a M1 too, :D). I would suggest to use Colab or any other external GPU