How To Use ComfyUI Nodes With SDXL 1.0 Base & Refiner Model (AI Tutorial Guide)

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
foreign how to use sdxl 1.0 and coffee UI workflow with sdxl notes hey guys I'm going to be showing you another tutorial for today we're going to be learning sdxl 1.0 and coffee UI workflow with sdxl nodes in the easiest way possible I'm also going to show you the difference between the output of sdxl base and sdxl refiner so without further Ado let's dive into the tutorial so first things first there are actually two ways for you to see a node so you can either right click and then add node or you can also click alt and then double click just like that so in this case I'm going to be using this one alt and then double click and then I will click checkpoint loader simple okay this one and then I'm going to change this one to sdxl base and then we're going to add a clip so again I'll double click and then this time I'm going to type sdxl alright so this one clip text in code sdxl so we're not going to use refiner yet so all right here we go and then I'm going to convert this one into input there you go there you go and then I'm going to be dragging this text G over here because I'm going to be adding a node and for utilities so primitive all right so this is where we're going to be typing the inputs for the positive and the negative okay and then connect that and I'm gonna copy this one for our negative the same thing here right there you go and then connect it to G and then to L all right and then I'm going to connect this one here in the clip the same thing here now the next thing that we're going to do is add a sampler so again alt and then double click and then I'm going to type sampler so I want to use the advanced this one and then I'm going to connect positive here and then our negative front over here all right there you go and then we're going to connect the model also to our checkpoint model and before we proceed I'm just going to type the prompt here as a real futuristic image of the world okay and here for a negative I just don't want to see a lot of people so I'm just going to type many people so I want this image to be focusing mainly on like how the world is going to look like according to this AI so again this is sdxl base 1.0 and then this time we're going to proceed so for the latent I'm going to just drag it here uh latent image I'm going to add empty latent image right over there and I'm just going to change the value of the height and the width so say for example 10 24 same thing here 10 24 and then I'm going to say for example generate about three pictures and then for the latent I'm going to add something here we're going to add vaed code and for the image right over here so you can either click save image or preview image if you want to click save image it will automatically save in your drive or in your local in my case I'm just going to click preview image because I want to see it first before saving it okay there you go and then let's put it here okay there you go and then and the next thing that I'm going to do is basically here so we're just going to change the sampler so it's up to you guys you can choose of course any of these sampler I'm just going to use this one and I don't have a GPU so this is going to fit better for me all right and yeah I guess that's it basically that's the only thing that we're going to do so we're going to load the image we're going to click here Q prompt let's see what sdxl base is going to give us oh yeah I forgot the vae of course yes there you go so you have to connect the vae all right so let's skew okay now it's running so you will see it running when it's turning green so that's when you know that it is processing all right so now we have generated three amazing images let's try to check them one by one okay uh there you go oh nice take a look at that so these are buildings and I'm guessing these are roads can you imagine this one wow look at that sphere so it's like a world or a city or country whatever covered in sphere oh nice yeah this is amazing these are all generated by AI you guys okay so what I'm gonna do is I'm gonna open this one in new tab for later so I've opened them in new tabs now we're not going to stop here we're going to see what sdxl refiner is going to do for us so the same thing click alt and then checkpoint loader this time I'm going to use refiner okay so I'm now using refiner so basically I'm just going to copy these I mean I'm going to copy this one okay there you go so I'm going to be using this for my positive as well and the same thing for the other one oops didn't mean to do that go back there here so this is for my negative so we're going to just connect this one okay I'm just going to borrow this one this is again for our positive and also this one for our negative so we're going to be using the same codes all right there you go so yeah I think we've connected it all right so for the sampler we're going to try to do the same thing so Advanced all right there we go and here we're just going to change a little bit our sampler so for this one we're going to use positive we're going to use negative and we're going to use the base model here so we've connected our positive and negative now this time I'm going to try to connect latent here so this is something that comes out after our basic and yeah we're going to proceed with this one so still the same vaed code all right and then the same thing we're doing want to make the same mistake as earlier we want to connect our vae and then here still the same preview image now before we are going to generate I just want to change something here so for this time we have here start that step and then we also have here steps number of steps we're just gonna reduce this one to say for example 10 all right okay and then for our refiner we are going to start here at 10 and then we are going to end that say for example 20. so that's what we are going to do and then still the same thing we're going to change the sampler name to the same one as earlier DPM plus plus underscore sde okay this one all right so yeah I think we have got it all so let's try to generate and see how different it would be so I'm going to Q prompt all right so there you go so it's starting here oops somehow I clicked it twice I didn't mean to do that I only meant to click it once all right but anyway it should be fine all right so now we have generated this is actually the second image that was generated because I've accidentally clicked here twice but let's see a difference in this case so let's close the old one so I'm going to open this one and we're going to open it again in the new tabs for us to have a comparison later on all right and the same thing here okay so now we're here so this is coming from base and then this is coming from refiner version all right so I'm going to show you the difference between the two so this is for the base as you can see there are yeah a little bit of details here but if you're going to go to a refiner you can see that there's a lot more details going on here say for example the building so if we are to click the building take a look at that one guys so here yeah there isn't a lot of things going on here it kind of looks like a mountain or something with this kind of particular building in the middle here you can see that there are really a lot of buildings going on here so you can see there's even like a bridge down here yeah this one doesn't look like a bridge at all so here it's very clear there are boats here ladies and gentlemen so yeah that's one of the difference another thing we're going to click on the second picture so as you can see here uh this is somehow like a UFO or something I think AI is thinking that they're going to be aliens in the future here in our world hopefully not okay but yeah this is how it looks and you can see that there are more details on the refiner version let's try to click the third one oh wow take a look at that that is definitely amazing and you can see the difference so yeah that's basically all there is that's how you're going to use sdxl 1.0 and coffee UI workflow with sdxl nodes and that's also the difference between the output between the base and the refiner version of sdxl 1.0 [Music]
Info
Channel: Future Thinker @Benji
Views: 12,257
Rating: undefined out of 5
Keywords: SDXL 1.0 tutorial, ComfyUI workflow with SDXL Nodes, Difference between SDXL Base and Refiner versions, AI image generation with SDXL, AI-generated surreal landscapes, Advanced AI image processing, SDXL 1.0 workflow guide, Exploring ComfyUI in Stable Diffusion, AI art with clear edges, Captivating AI image results, stable diffusion, stable diffusion tutorial, comfyui, comfyui nodes, comfyui tutorial, stable diffusion xl, stable diffusion ai, ai art, ai, sdxl nodes, sdxl 1.0
Id: xiZPd93Hwg4
Channel Id: undefined
Length: 9min 40sec (580 seconds)
Published: Wed Aug 02 2023
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.