Get Better Results With Ai by Using Stable Diffusion For Your Arch Viz Projects!

[@MatteoFontana]

Best Stable Diffusion video for Archviz! Only the essentials, explained precisely but calmly without overwhelming the viewer.

[@max22able]

Thanks for this - I've made it through to the end and can get going with AI. Just a quick update with the latest version - at 14:30 where you drag the updated image back into inpaint, this is no longer a drag/drop operation but is done with a "send to" button under the generated window. Thanks again.

[@TheGalacticIndian]

Cieszy, że kanał na bieżąco z nowościami👌Brawo!

[@ArchVizArtist]

Dzięki ;)

[@ameralhomsi8847]

you did a lot of work to make this video! thank you so much

[@py8115]

Hi, Ava, thank you for the tutorial it is the only one in architecture field in English! Thank you a lot:) would be super interesting to get to know your experience with models, Lora, negative models. Maybe it is idea for another video:) One more thing I saw during your process with deliberate model, you will not need to blend some parts in photoshop if you will use inpainting version of model (it usually placed in very bottom of description to model in civitai) All the best! Pavel.

[@ArchVizArtist]

Great suggestion!

[@9JANOOB]

Hey, Fantastic Video. I highly recommend that python be downloaded from the Microsoft Store. I tried downloading it directly from the website and ran into some issues but none when downloaded from the microsoft store. Also, I think you may have forgotten to put the information about the upscaler you adviced in the blogpost as you said in the video :).

[@gp4f415]

Very true! i tried from the blog and it didnt let me pass the 5º step, your comment helped and i could install the webui-user.bat file! Thanks

[@CrazyMan-qs4so]

Working if downloaded from Microsoft Store. Thank you!

[@atajqtsawa3293]

Super, że przekładacie najnowsze świeżynki na realny grunt. Właśnie skończyłem kurs obsługi SD i to jest niesamowite! Wpadłem w wir poprawiania swoich wizek :)

[@ArchVizArtist]

Dzięki za komentarz ;)

[@shinonyx]

Man, i would use this just for people, the result was amazing

[@ArchVizArtist]

It is :)

[@david.rosyada]

Thanks Ava!! really useful for an AI Newbie..btw, your render is awesome already...!!

[@aang_arang]

doesnt seem to work, I keep getting a multitude of errors almost immediately. ERROR: THESE PACKAGES DO NOT MATCH THE HASHES FROM THE REQUIREMENTS FILE. If you have updated the package versions, please update the hashes. Otherwise, examine the package contents carefully; someone may have tampered with them.

[@hotlineoperator]

You managed to get a lot of information into a compact package. At the moment, I am interested in how AI becomes part of the 3D workflow, like - how simple a 3D model can be used so that it can be used to generate an AI image - how AI can make textures that are used in 3D modeling Regeneration of parts of the image can also be done directly in Photoshop, but it cannot do everything that SD can do - it depends on the model and settings. I'm really looking forward to it that in the future, AI will learn to make 3D models directly as well, it will be interesting.

[@ArchVizArtist]

Thanks for sharing your thoughts ;)

[@pankajteli2746]

so much to learn bout AI from this one video! thank you Aga! 😊😊

[@ArchVizArtist]

My pleasure!

[@Lewhateverr]

Hi Ava, you are awesome for this tutorial and all other content! Thanks so much! Following these steps should I just download models that have SD 1.5 in the ''base model'' description? There are others like DXL Turbo or SDXL Lightning (no idea whats all this, newbie here in IA)

[@puja1985]

If you want to showcase your graphics card's capabilities or promote it, stable diffusion may be the best choice of a tool. Most people don't have a powerful CPU or GPU and prefer to use GPU cloud services that also provide AI image generation tools, like Leonardo AI. Leonardo AI is also a stable diffusion-based model, designed by developers in a way that artists don't need to worry about technical details such as multiple installation steps or potential errors during testing. By using Leonardo AI, it saves me from technical headaches and allows me to focus on being more creative. Every feature mentioned in this video is also available in Leonardo AI, with the additional cost of their GPU services. These features include text-to-image, image-to-image, inpainting, outpainting, etc.

[@MegaLauritagarza]

Esta muy bueno el tutorial! 🔥Un poco largo pero vale la pena con todos los temas que se ven.!!👏

[@sarainr2115]

THANK YOU!! You cleared out a lot of things to me !

[@anamariac5775]

Hi Ava! I appreciate your tutorials, they are very helpful. Could you make a tutorial showing us how to make a boucle material that looks as realistic as possible and that can be in different colors (for example a gray boucle, a green boucle, etc.)? Thank you!

[@ArchVizArtist]

Thank you, maybe next year :D I have my calendar already full for this year ;)

[@АлександрСергеевич-с4х]

Lol, use dsp with noise map lol

[@marweiUT]

Díky!

[@Lustigorn]

Today is a really useful and interesting video. many thanks👍

[@ArchVizArtist]

Glad you liked it!

[@RVArq]

Thanks a lot. It Washington very helpfull specially the part where u work with 5 k images

[@mae2309]

where are the link to the checkpoints and the upsacle? amazing video BTW! thank you cheers

[@tomaskorch6580]

Hi, great video. I noticed that at vividvisions, they are using stable diffusion 2.1. Is that different from 1.5? Or is just a checkpoint model that is 2.1. Im bit confused. Can someone please explain differences? Thank you :-)

[@vitoorvieeira]

Greatt!! do more tutorials of stable diffusion for archviz

[@elianaalvarez5954]

awesome tutorial! Thank you so much!!

[@vapa86]

can you suggest a good architecture/archviz checkpoint for stable diffusion? Thanks

[@GiorgiBekurashvili]

One of the most useful tips here

[@damianlazo5760]

amazing! can you make a updated video like this?

[@abderrahmenezzine]

Thanks for sharing This knowledge , very very much appreciated

[@ArchVizArtist]

My pleasure!

Have you found out if AI could be used to denoise our arch viz images? For example, Arnold can denoise images, but LPEs or AOVs don't or it is quite complicated and time consuming. AI could help with denoising, don't you think?

[@a.miroshin]

Thank you!

[@VladYakovenko-q6q]

I'm interested, in the video Vivid studio we can see that they use high resolution in inpaint models. They use sd 1.5 model? How then does it work for them? I'm trying on the SDXL model and I can't good results on it. For example areas with grass turn out terrible. Sometimes he makes bushes smeared. It is advisable for a lilac bush to write in promt that this lilac then it does well. Maybe you are not showing something important in video? Everything is clear with people, but with plants? I was only able to get good results in the background and sometimes in the middle ground 🙂

[@ArchVizArtist]

I have shown the whole example with planting at the end...

[@omargamal7858]

first thank youu!!..........but i have an issue with this Step 3: Download the Stable Diffusion 'git' is not recognized as an internal or external command, operable program or batch file.

[@davidsolano2423]

Wow!!!! Thanks for Sharing!

[@ArchVizArtist]

Our pleasure!

[@cgimadesimple]

great video!!

[@johnny5132]

hey there, thanks for this, however i cant see the download link for the 4x ultrasharp?

[@m.g6991]

Hi! the link on step 4 doesn't work! for me :(

[@miki1510]

I legit freaked out watching this video. I'll explain myself. I recently started using AI to generate images. I mainly use them to sell posters and T-shirt on many different websites. I know the potential of AI generated images and videos. And while watching this video something wrong clicked in my brain: am i watching a real woman teaching me AI stuff or is it a 50 years old fat guy from his basement programming an AI girl and abusing the YT algorith for views? Not that it would really matter to me, i still learned something. Luckily, by checking your YT channel, i found out 3 years ago you couldn't make such a good AI generated video and i'm 99,99% you're a real woman/person. And you might think i'm being paranoid. You're absolutely right, but this might be our future in 4-5 years at the rate this technology is evolving. WHAT HAVE WE DONE?!?!?🤣🤣🤣😬

[@ArchVizArtist]

Hehehe

[@vrendviz1886]

hi nice tutorial i saw the studio tour of Vivid Vision from your channel and that they explaining about populating their scene with SD but didnt expect that you will be giving us a great tutorial about it by the way i followed the steps explained in your website and Stable diffusion failed to load

[@arjunsharma390]

Hi when I run the webui-user batch file I get this error stderr: Python was not found; run without arguments to install from the Microsoft Store, or disable this shortcut from Settings > Manage App Execution Aliases. I have installed python? any idea what to do

[@박명수-n2m]

How did you create the images used in the "Improve Greenery with Inpaint" section?

[@ArchVizArtist]

These are my 3D renders, not AI generated images.

[@박명수-n2m]

@@ArchVizArtist thanks, is it D5?

[@FerryPasschier]

Are you sponsored by Nvidia? Because it also works perfectly fine on my not-Nvidia card

[@ArchVizArtist]

Yes it's sponsored, but the part that you need the NVIDIA card was reasearch, many blogs says you need that, e.g. gigazine.net/gsc_news/en/20220907-automatic1111-stable-diffusion-webui/ Now I'm confused :D, it doesn't matter too much though.

[@ericpanzer8159]

@@ArchVizArtist It does matter because you provided people with incorrect information that will direct them solely toward the GPU brand that was the sponsor of your video. It is true that Nvidia cards are (for now) the fastest and most flexible when it comes to using SD. But for someone on a budget and who also wants a GPU with good gaming performance, AMD may offer more cost-effective options. This video was also promoted as being about using it for architectural visualization, but you scarcely touch on that.

[@hansrama3485]

​@@ericpanzer8159Its clearly stated that its sponsored. what more do you do. Nvidia does have the best cards though doesn't it even if their pricing can be fishy you dont always need the latest cards and yes they make the best cards no argument there

[@abubakarabdul9755]

This is really great Ava, I really appreciate your time and effort you put in in your tutorials… Please can you just do a quick tutorial on how to simulate fountain using chaos phoenix??? I have tried the methods on youtube but it doesn’t seem to simulate on my ends😢… My pc = msi creator, 32gig ram, rtx4060

[@ArchVizArtist]

Sorry to hear that, maybe next year, my calendar for this year is fully packed ;/

[@HANKUS]

i would try doing the water in stable diffusion. photoshop or draw some water onyour 3d rendering and then let Stable diffusion improve it. if you send me a jpg i can try as im curious myself

[@raulkassab25]

Step 8 is not working: Error: Already up to date. '#automatic' is not recognized as an internal or external command, operable program or batch file.

[@guyberryman5226]

hi thank you so much for this tutorial but I am faced with the error expecting value: line 1 column 1 after on step 5

[@UmbralArq]

couldn't set the step 5, any solution?

[@АлександрСергеевич-с4х]

If I have GeForce RTX 3060 12GB (!BUT FROM ASUS, NO nvidia!), will it work?

[@ArchVizArtist]

Yes

[@MostafaMohamed-vq9oi]

Hi AVA , can you share the link of model that you used in stable diffusion to get this high resolution ?

[@DaNiElMmM93]

its only possible with nvidia video card? :(

[@pedrolacueva]

don't know why but I am not able to install the webui-user. I click but nothing happen :( windows 11 Pro, 64 gb ram. rtx 4090

[@LucyHarutyunyan]

Thank you for video, i install it, all good, but its not generate becouse of error connection, but why? I have connection but it not genarate.

[@tami_boo]

Thank for this useful video! I have RTX 4090, but it is slow(( maybe there are some settings to male it fast as yours?

[@ziyuanime]

mine 4070 and its quite fast

[@imflaviopires]

Amazing!

[@ArchVizArtist]

Thank you! Cheers!

[@BobbyWParkerJr]

A lot of work to get it running. I doubt many will try. If they come up with a simple installer, that would be great.

[@BobbyWParkerJr]

I installed and it isn't too bad. The results are awful, so I downloaded some models, but they are safetensors. How are you converting them to ckpt?

[@poison4u]

Where to download stable diffusion model?

[@vitoorvieeira]

i pressed ctrl+c to copy the link and then somethign happened, how can i fix it?

[@EcclesiaToday]

Thank you so much!

[@ArchVizArtist]

You're welcome!

[@DanFrederiksen]

Very fast but maybe not as capable as midjourney. but fancy it can run locally. Are there no bigger models that 2GB? maybe 10 if a 12GB card can handle that

[@leonardoaguilerahonduras2067]

Prome ai + PS + Krea ❤ 🎉

[@johnny5132]

hey who did the studio tour? sorry i didnt catch the name and i couldnt see it in the description?

[@rameshkumar.k7133]

It's really useful

[@ArchVizArtist]

Glad to hear that

[@L3nny666]

nice...would be even better if it wouldn't have the resolution restriction.

[@ArchVizArtist]

Would be great ;)

[@renanarchviz]

Couldn't launch python exit code: 3 stderr: The system cannot find the path specified. this error happened

[@josedanielfernandezmatheu620]

the same to me

[@शुभचिंतक]

Step 4: Download a checkpoint model Link not working. Please someone help me @ it

[@minkim6010]

is it possible to run stable diffusion in mac os?

[@raulkassab25]

When I try to open the web-user.bat after copying and pasting the edited tab, this is what I get, and no URL is shown: Already up to date. '#automatic' is not recognized as an internal or external command, operable program or batch file. Any help regarding this topic? Thanks

[@bouidiouamohamedlamine3766]

same issue :(

[@asilbekabduqodirov2935]

👏👏👏👏

[@fouezdesign]

Great tutorial but you lost me with the machine specs...

[@syahti]

How to Load Safetensors ?

[@juanteran5724]

Step 5 not working for me. Message "Stable diffusion model failed to load" showing at the end.

[@josedanielfernandezmatheu620]

x2

[@simonfinger8760]

same 😭

[@simonfinger8760]

fixed. Forgot to click the checkbox 'add python 2.10 to PATH'. Reinstalled. Working now.

[@andree839]

Hi! I really want to enable the setting for saving propts as text files but for me the setting is nonexistent in the settings tab. Do you know if it has been placed somwherer else?

[@andree839]

So I discovered you can access this information in the "PNG info" tab i Stable diffusionwebUI but it would be useful to also see it directly in the output folder.

[@simonfinger8760]

I can't get past step 5 either 😭. Python not found - launch unsuccessful 😖

[@simonfinger8760]

fixed. Forgot to click the checkbox 'add python 2.10 to PATH'. Reinstalled. Working now.

[@Mohamed-Al-Masry]

Very nice 👌 ❤

[@ArchVizArtist]

Thanks 🤗

[@itsweird9858]

is stable diffusion is free to use?

[@efezeybek4226]

please make more ai videos

[@justkater]

Nice

[@ArchVizArtist]

Thanks

[@MrWanted000]

Would it not be easier to use Photoshop AI for that type of modification?

[@ArchVizArtist]

Not at all, PS gives random results, here we can control how similar to the original image we are. Last example is almost the same as my render, it wouldn't be possible with PS AI.

[@filipcvetkovic1993]

Is this free? Why not provide the link to buy if not free?

[@AI_Arc]

Freeeeee

[@RaulGarcia-oh6gd]

error code 1

[@rdxblaster9051]

first comment 🥰

[@ArchVizArtist]

Yes you are! Thank you!!!

[@bryanbrewer4272]

Kreamy

[@Alex29196]

Your voice sounds generated by 11eleven labs. Sorry

[@erickbooster1]

Sounds like she live in china and speaks mandarin as her first language

[@PivaldiEnjoyer]

This is a real voice, there is no such high-quality lipsink yet

[@juty6852]

I tried installing it last year. I was disappointed with results. Very low level of quality, variation, realism, everything is very mediocre and weak.

[@astonyo7544]

best of nothing!

[@jameshill1243]

This is so depressing. 😔

[@hansrama3485]

why? honestly want to know

[@puja1985]

​@hansrama3485 maybe depressing bcs theres just so much installation process to go through..

[@HANKUS]

absolutely not, the Ai shown in this tutorial only enhances the work you give it. saves you time too (people blending) its still a hassle to learn, but its no different than filters. you can take a garbage photo and give it a great photo and it is still garbage, give it a good photo and you can make it better.... but the base comes from an artist

[@heikohesse4666]

What a shame, I thought you gave some good Arch Viz tips - how the whole thing works and how to use it has already been explained better in a thousand other videos...

[@astonyo7544]

please stop this nonsless tutorials trying to make a profit - it just doesn`t work!

[@syahti]

'#automatic' is not recognized as an internal or external command, operable program or batch file. Already up to date. venv "C:\Stable Diffusion\stable-diffusion-webui\stable-diffusion-webui\venv\Scripts\Python.exe" Python 3.10.6 (tags/v3.10.6:9c7b4bd, Aug 1 2022, 21:53:49) [MSC v.1932 64 bit (AMD64)] Version: v1.6.0-2-g4afaaf8a Commit hash: 4afaaf8a020c1df457bcf7250cb1c7f609699fa7 Installing xformers me eror for step 8

[@BernardoBenítez-h3v]

did you found how to fix it?