---
title: 'Getting Started with Stable Diffusion in 2024 for Absolute Beginners'
source: 'https://youtube.com/watch?v=mcqREz9xE74'
video_id: 'mcqREz9xE74'
date: 2026-06-30
duration_sec: 776
---

# Getting Started with Stable Diffusion in 2024 for Absolute Beginners

> Source: [Getting Started with Stable Diffusion in 2024 for Absolute Beginners](https://youtube.com/watch?v=mcqREz9xE74)

## Summary

This video provides a step-by-step guide for absolute beginners to install and run Stable Diffusion locally on their own machine. The creator covers downloading Python, obtaining a Stable Diffusion model, and setting up the web UI interface to generate AI images.

### Key Points

- **Introduction to Stable Diffusion** [0:00] — The video introduces Stable Diffusion as a popular text-to-image AI model that can be run locally without limitations.
- **Prerequisites: Python Installation** [1:48] — Python is required to run Stable Diffusion. Download the latest version from python.org and ensure it is added to the system PATH.
- **Downloading a Stable Diffusion Model** [2:39] — Models are available for free from stability.ai. The recommended model for this tutorial is SDXL Base 1.0, which is about 6 GB.
- **Downloading the Web UI** [6:19] — Search for 'stable diffusion web UI' on GitHub and download the ZIP file. Extract it to a local folder.
- **Running the Web UI** [7:41] — Execute the 'webui-user.bat' file (Windows) or 'webui-user.sh' (Mac/Linux) to install dependencies and launch the interface. First launch takes longer.
- **Selecting a Checkpoint and Generating an Image** [8:46] — After launch, select a checkpoint (e.g., the default v1-5 model) and enter a prompt like 'cute sleeping cat outdoors quaint porch evening light' to generate an image.
- **Installing the SDXL Model** [9:31] — Move the downloaded SDXL model file into the 'models/Stable-diffusion' folder inside the web UI directory. Refresh checkpoints and select the new model.
- **Adjusting Resolution for SDXL** [10:29] — SDXL works best at higher resolutions like 768x768. Change width and height in the interface for better results.
- **Refining Prompts and Final Tips** [11:17] — Adding keywords like 'photo realistic' improves output quality. The video notes that generated images may have imperfections (e.g., missing legs) that can be fixed with inpainting or other techniques.

### Conclusion

Stable Diffusion is a powerful, free tool for generating AI images locally. With the steps outlined, beginners can set it up and start experimenting, though further refinement and prompt engineering are needed for optimal results.

## Transcript

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