[0:00] hey everyone and welcome back to the
[0:01] channel in this video I want to show you
[0:03] how you can start generating your own AI
[0:05] images using stable diffusion which runs
[0:08] locally on your own machine so you can
[0:10] run it at will however often you want
[0:12] without any limitations and if you don't
[0:14] know what stable diffusion is you may
[0:15] have been living under a rock stable
[0:18] diffusion is one of the currently most
[0:19] popular text to image AI based models
[0:22] that you can use to generate AI images
[0:25] whether you call it AI art or not is a
[0:27] whole another discussion maybe a topic
[0:29] of another video that I'll be making but
[0:31] you can generate some really really
[0:32] amazing photo realistic or artistic or
[0:36] you know just Creative Images for all
[0:38] sorts of things using stable diffusion
[0:40] I've been using it for a lot of my own
[0:41] wallpapers that I've generated just you
[0:43] know just because I wanted something
[0:44] nice on my desktop that I couldn't just
[0:46] get online as here's one for Japan you
[0:49] can also generate you know images of
[0:51] cats of cityscapes of people of you know
[0:53] monsters cuz I like grimy and horror so
[0:55] I've done a whole bunch of that sort of
[0:57] stuff with stable diffusion the other
[0:59] thing I I've used stable diffusion for a
[1:01] lot as well as to generate some concept
[1:02] images for a video game I've been trying
[1:05] to create so there's a whole bunch of
[1:06] images generated by stable defusion it
[1:08] has a lot of really powerful features
[1:10] you can you know use input images to
[1:11] generate images that are like the images
[1:13] that you feed into it it now supports
[1:15] text to video as well as video to video
[1:17] and a whole bunch of other Advanced
[1:18] features but in this video I just want
[1:19] to go over the absolute basics of how do
[1:21] you get started setup install stable
[1:23] diffusion and how do you start running
[1:25] and generating your own images on your
[1:26] own computer so I want to run through
[1:27] all of that in this particular video
[1:30] timestamps and everything so you can
[1:31] jump to what you might find relevant or
[1:33] interesting will be down in the video
[1:34] description YouTube also adds the
[1:36] chapter so you can just jump to wherever
[1:37] you want to and any questions you might
[1:39] have comments feedback leave that down
[1:40] in the comment section I'm pretty
[1:42] proactive reading everything and try to
[1:44] answer as much as I can so just utilize
[1:46] that as much as you want to now in order
[1:48] to run stable diffusion you do need a
[1:50] couple of things the first thing you
[1:52] need to do is you need to download
[1:54] python python is a programming language
[1:56] and stable diffusion runs on python so
[1:59] any machine that you can install python
[2:01] on be that Windows Mac Linux you can run
[2:03] stable diff Fusion on so if you jump
[2:05] over to python.org downloads and I'll
[2:08] drop all of the links down in the video
[2:09] description as well as well as a little
[2:11] you know step-by-step instruction of
[2:12] what you need to do to get up and
[2:14] running as well just jump to python.org
[2:16] downloads and then just simply cck
[2:18] download python 3.1 2.1 or whatever the
[2:20] latest version of that might be and then
[2:23] just save that out your local computer
[2:24] and then all you have to do is run that
[2:26] installer to make sure that you add
[2:28] python to the path because you still
[2:29] need to execute it from the command line
[2:31] but again VI scripts and I'll show you
[2:33] how to do that later it's not difficult
[2:35] it just it sounds scary first just
[2:37] install python onto your machine then
[2:39] that is done the next thing you need to
[2:41] do is you need to download a stable
[2:43] diffusion model which is essentially the
[2:46] AI built model that contains all of the
[2:48] knowledge of how to generate the images
[2:51] now there's a bit of misconception that
[2:53] these models contain copies of images
[2:55] that they found online which isn't
[2:56] really the case it's more like imagine a
[2:59] a robot looking at every single image
[3:01] within a very big image database and
[3:04] remembering what's the shape of a cat
[3:05] what's the general shape of a human or
[3:07] house or tree and so that when you ask
[3:09] the AI to generate me a tree it just has
[3:12] the knowledge of how generally to shape
[3:14] up a tree based on all the images it
[3:16] seen it's kind of like drawing from
[3:18] memory now that is not to say that there
[3:20] isn't you know legal issues and
[3:22] copyright questions and a whole bunch of
[3:23] things that people are still trying to
[3:25] figure out a lot of controversy around
[3:26] that and again might be the topic for
[3:28] another video if you guys are interested
[3:30] but you can get these stable diffusion
[3:32] models for free online and the easiest
[3:34] place to get them is just come to
[3:35] stability AI which is the company that
[3:37] makes and releases stable diffusion and
[3:40] by the way stable diffusion is open
[3:41] source so the source code for stable
[3:43] diffusion is freely available online you
[3:44] can look at it you can download it you
[3:46] can modify it it's all free in that way
[3:48] and so are the models that stability AI
[3:50] releases so come to stability. a and
[3:53] then you can come down and you can look
[3:54] at the different models and all of these
[3:56] AI models excel at something specific
[3:59] the latest one is sdxl Turbo which is a
[4:01] really fast version of the stable
[4:04] diffusion model but it you know Works
[4:05] slightly different to what you'll find
[4:07] mostly online so the one I want to go
[4:09] with for this video at least is stable
[4:11] diffusion XL and you can see some of the
[4:13] example images that have been generated
[4:15] using this model and it's it's just
[4:17] mind-blowing what you can do with this
[4:18] so let's close this out and down here
[4:21] now you can generate images online
[4:22] without running anything on your
[4:24] computer but you will have to pay for
[4:25] that privilege on a Pro Plan for
[4:27] stability AI we actually just wanted to
[4:29] downloaded the model and run everything
[4:30] locally on our machine so let's click on
[4:32] download code and this will open up a
[4:34] GitHub repository which is essentially a
[4:36] a place where developers can place code
[4:39] and you know technical stuff don't worry
[4:41] too much about it let's just scrub all
[4:43] the way down skip over all of the news
[4:45] and all of the other bits and pieces
[4:47] what you looking for are these links to
[4:48] sdxl base 1.0 there's also a refiner
[4:52] which is kind of like a second pass over
[4:53] your image to improve some of the
[4:55] details and hands and feet and enhance
[4:57] the image we're not going to worry about
[4:58] that at least not for this particular
[5:00] video so simply click into this sdxl
[5:02] base 1.0 and again don't worry I'll I'll
[5:05] put all of the download links down below
[5:06] the video and then you're going to come
[5:08] to huggingface doco which is a site that
[5:11] houses a lot of the stable diffusion and
[5:13] AI text to image or text to video models
[5:17] and information um and in here this is
[5:19] the one for stability Ai and stabil
[5:21] diffusion XL the base model 1.0 again
[5:25] just ignore all of the technical stuff
[5:26] but it is actually really interesting if
[5:28] you want to know how these models work
[5:29] again I can make a separate video on
[5:30] that read through some of the text specs
[5:32] it's actually really exciting I think on
[5:34] how these models actually work or if
[5:37] none of that is of interest you you just
[5:38] want to generate some images just come
[5:40] into files and versions up here at the
[5:42] top come down a little bit and the one
[5:44] you want to download is sdxl base 1.0
[5:47] vae which stands for variable Auto
[5:49] encoder again just a part that improves
[5:52] the image as it comes out but it's all
[5:53] baked into this one model here and let
[5:56] just click on this download file Link at
[5:58] the base then let's download this file
[6:01] now it's pretty massive some of these
[6:02] models are you know many gigabytes this
[6:04] one is 6 GB I'm going to let that
[6:06] download in the background while we get
[6:08] set up with the rest of the stuff now
[6:10] we've installed python which is the
[6:11] runtime that's going to run stable
[6:13] diffusion and we are downloading the
[6:14] model that stable diffusion will use to
[6:17] generate our images we do need stable
[6:19] diffusion itself however and the way to
[6:21] get that simply come into Google and
[6:23] search for stable diffusion UI and again
[6:25] another GitHub code repository called
[6:27] stable diffusion web UI and again links
[6:29] down in the video description and this
[6:31] is the actual code the actual executable
[6:33] code that runs stable diffusion with a
[6:35] nice little webbased interface that you
[6:37] can then enter your text in and run your
[6:39] prompts and you know look at your images
[6:40] and everything like that so up here if
[6:42] you have git installed if you know what
[6:44] that means you can just download the
[6:46] GitHub repo you can just clone it to
[6:47] your local machine if you don't know
[6:49] what git is you don't really care up
[6:51] here where it says code just click on
[6:52] this little drop down here and we're
[6:54] just going to say download zip so we can
[6:56] bypass all of the deie things and just
[6:58] download the whole thing thing it's
[6:59] going to download stable diffusion webui
[7:01] Master zip to our local machine so let's
[7:03] do that back in the folder where you
[7:06] downloaded everything so there's our
[7:07] stable diffusion webui Master zip file
[7:10] let's unpack all of that here why not
[7:13] into stable diffusion web UI master so
[7:16] let's jump into this particular folder
[7:19] okay a sub folder doesn't really matter
[7:20] too much and in here is stable diffusion
[7:24] everything that you need to run it as
[7:26] well as add the models that you want to
[7:27] use to generate your images or tools for
[7:30] modifying images or trending your own
[7:32] models everything sits in here and again
[7:34] if you're a Dev or an engineer there's
[7:36] some fascinating stuff in here worth
[7:37] looking through and understanding how it
[7:39] all hangs together if you're not the
[7:41] easiest thing you need to do is you need
[7:42] to execute web user UI batch file if
[7:44] you're on Windows or web user UI shell
[7:47] file if you're on Mac or Linux so I'm on
[7:50] Windows I'm just going to double click
[7:51] this web UI user batch file and that is
[7:55] going to now kick off and execute Python
[7:58] and install all of the dependency all of
[8:00] the packages and all the other bits and
[8:01] pieces that I need to run stable
[8:03] diffusion locally now this might take a
[8:06] while because it's going to download
[8:07] quite a lot of data run through a lot of
[8:09] different tools and get everything set
[8:11] up and then it's going to launch the web
[8:13] UI that we're getting which will then
[8:15] allow us to enter our prompts and run
[8:16] images as well as use our different
[8:18] models to generate the images so I'm
[8:20] going to let this run you may see me
[8:22] just hop out a little bit while this
[8:23] runs because again this this can take a
[8:25] little bit so I'm just going to Tinker
[8:27] while this
[8:28] runs okay here we are and don't worry
[8:30] this only happens the very first time
[8:32] you launch stable diffusion from here on
[8:34] it usually starts up within 10 to 20
[8:36] seconds so it's just the first time that
[8:37] this is really slow so now stable
[8:39] diffusion has installed all the
[8:40] dependency and it has launched the webui
[8:42] which is where you can do everything
[8:44] with stable diffusion that you may want
[8:46] to do now the first thing you need to do
[8:47] if you want to generate any images with
[8:49] stable diffusion is you need to select a
[8:50] stable diffusion checkpoint by default
[8:53] if you click on this drop down here
[8:54] there's V 1.5 pruned EMA only safe
[8:57] tensil it's like a dummy model but it's
[8:59] enough to test whether this whole thing
[9:01] is even working so let's simply in this
[9:04] prompt field here right uh cute sleeping
[9:08] cat
[9:10] Outdoors quaint porch evening
[9:15] light and then let's hit generate over
[9:18] on the right hand side and see what
[9:21] happens and show enough it looks like a
[9:24] cat sleeping outside on the porch so
[9:26] this seems to be working however the
[9:29] default model isn't terribly great so
[9:31] let's come back to our download folder
[9:33] and yes the sdxl base 1.0 safe tenses
[9:37] file has downloaded and it's 6.7 GB so
[9:40] that took a little bit so let's cut this
[9:42] out and we need to move this into the
[9:43] stable diffusion webui Master folder and
[9:47] it goes under models and in here you'll
[9:49] find stable diffusion so double click
[9:51] into that and paste or move your model
[9:54] into that let's come back to the webui
[9:57] hit Refresh on those checkpoints
[9:59] and now in this drop down you should
[10:00] have sdxl Base 1.0.9 VA let's click on
[10:04] this and load it up it's it's going to
[10:06] take a little while while this model is
[10:07] being loaded and if you pull up the
[10:09] console where you've launched the web
[10:11] you can see what's happening down here
[10:13] it's now loaded this model let's come
[10:16] back to stable diffusion repap and hit
[10:17] generate again to regenerate that image
[10:19] using a different
[10:22] model and yeah that is still a cat on
[10:26] the porch outside at evening light now
[10:29] with stable diffusion XL you should
[10:32] really be changing your resolution so
[10:33] right now the resolution of our output
[10:35] is set to 512 x 512 which used to be the
[10:38] resolution at which some of the earlier
[10:39] models for stable diffusion were trained
[10:41] so they trained on images at this
[10:43] resolution which meant the outputs were
[10:44] best using that resolution stable
[10:46] diffusion XL however and some of the new
[10:48] models use higher resolution so I highly
[10:50] recommend change your width to 768 by
[10:54] 768 obviously that's going to take
[10:56] longer to generate depending on your
[10:57] graphics card and stable diffusion does
[10:59] require a fairly decent graphics card I
[11:00] would recommend at least 4 GB of vram
[11:03] the Nvidia RTX cards work really really
[11:06] well for this so let's hit generate and
[11:08] regenerate that
[11:11] image and yeah it's still a cat on a
[11:13] port outside don't like that it's kind
[11:15] of cartoony looking so let's refine our
[11:17] prompt just a little bit let's add photo
[11:19] and realistic into our prompt and let's
[11:22] regenerate our
[11:24] image and that looks pretty cool you can
[11:26] click on this image to zoom in and you
[11:28] can see the iess and the fact that it's
[11:30] generated and the fact that this cat
[11:32] seems to be missing a leg this is just
[11:34] some of those things that you'll have to
[11:36] encounter and you'll have to fix them
[11:37] manually either with in painting or with
[11:39] some other trickery to kind of get them
[11:41] to look just right but let's just come
[11:43] back let's just maybe just generate one
[11:44] more so you can see that every single
[11:45] time I get a new image generated that
[11:48] matches my prompt and there is a lot to
[11:51] how you do the prompting both the
[11:52] positive as the negative prompt some of
[11:54] these parameters they all influence that
[11:56] final image but that's beyond the scope
[11:58] of this video
[11:59] from this point just play around and
[12:01] just have some fun and see what you can
[12:03] get there's some amazing stuff you can
[12:04] do and I'll cover all of these
[12:06] parameters and how to prompt properly as
[12:07] well as some more advanced prompts and
[12:09] other stuff with stable diffusion in
[12:10] later videos and again if you have
[12:11] questions you want to see something
[12:13] specific or you're just getting stuck
[12:14] somewhere just drop me those comments
[12:16] down below and I'll get back to you I'm
[12:18] just I'm really excited I really like
[12:19] stab diffusion and some of these other
[12:21] cool tools that are coming out
[12:22] everywhere with AI everything um even
[12:25] though there are you know questions
[12:27] around copyright impact of eye on the
[12:29] workforce and how everything is Shifting
[12:32] know there's a lot to unpack don't want
[12:34] to get too much into it but it is a
[12:35] really cool tool to play with and I've
[12:37] used it for tons of my own personal
[12:39] stuff and just have fun with and play
[12:41] around but with that I'm going to wrap
[12:42] up this video If you enjoyed it please
[12:44] hit that like button subscribe to this
[12:45] channel if you want to support me it
[12:47] really helps me out and I greatly
[12:48] appreciate it and with that thank you
[12:50] very much for watching and until next
[12:51] time I will see you
[12:55] later