How to Install and Use ComfyUI and SwarmUI on Massed Compute and RunPod Private Cloud GPU Services
Full tutorial link > https://www.youtube.com/watch?v=bBxgtVD3ek4
If your GPU is not strong enough to run Generative AI models this is the tutorial that you need. Or you want to scale your generation speed by using multiple GPUs at the same time again this is excellent tutorial. In this tutorial I will show how to setup ComfyUI and SwarmUI literally 1-click on Massed Compute and RunPod and use your most liked best image and video generation models like Qwen, FLUX, Wan 2.2 and more.
🔗 Important Links:
Download ComfyUI Installer: https://www.patreon.com/posts/105023709
Download SwarmUI Installer and Model Downloader: https://www.patreon.com/posts/114517862
Previous Detailed Windows Tutorial (Recommended Watch): https://youtu.be/c3gEoAyL2IE
⏰ TIMESTAMPS / CHAPTERS
00:00:00 Introduction & Tutorial Goals
00:00:39 Downloading the ComfyUI Installer & Reading Update News
00:01:06 Downloading the SwarmUI Installer & Checking Changelogs
00:01:25 Extracting ComfyUI & Opening the Massed Compute Instructions
00:01:48 Deploying on Massed Compute: GPU Selection
00:02:03 Applying the 'SECourses' Coupon Code
00:02:30 Choosing a Multi-GPU Machine for the Demo
00:02:55 Installing the ThinLinc Remote Desktop Client
00:03:12 Crucial: Configuring ThinLinc Local Devices & Shared Drives
00:03:52 Connecting to the Massed Compute Desktop
00:04:13 Transferring Installer Files to the Remote Machine
00:05:18 Installing ComfyUI via Terminal Command
00:06:31 Updating the Pre-Installed SwarmUI on the Machine
00:07:00 Preparing the SwarmUI Model Downloader
00:07:48 Launching the Downloader & Downloading Model Bundles
00:08:46 Launching SwarmUI with a Public Cloudflare Link
00:09:11 Configuring the First Backend (GPU 0) with Sage-Attention
00:10:04 Adding and Configuring the Second Backend (GPU 1)
00:10:33 Importing the 'Amazing Swarm' Presets
00:11:04 Live Demo: Generating Realistic Images
00:12:48 Monitoring the Multi-GPU Generation Process
00:13:32 How to Download Your Generated Images (Two Methods)
00:14:12 IMPORTANT: How to Stop Billing by Deleting the Machine
00:15:07 Part 2: Starting the RunPod Installation
00:15:26 Deploying a RunPod Pod: Choosing the Right Template
00:16:24 Setting Pod Volume Size and Overrides
00:16:41 Troubleshooting: Handling a Pod That Won't Start
00:18:07 Uploading & Installing ComfyUI on RunPod
00:19:30 Uploading & Installing SwarmUI on RunPod
00:20:34 First-Time SwarmUI Setup Wizard (Important Settings)
00:21:04 Configuring Multi-GPU Backends on RunPod
00:22:14 Downloading Models Using the SwarmUI Downloader on RunPod
00:23:51 Importing Presets into SwarmUI on RunPod
00:24:31 Live Demo: Generating Images on RTX 4090s
00:25:52 Downloading Your Images from the RunPod Workspace
00:26:37 RunPod Billing: Stopping vs. Terminating Your Pod
00:27:16 Conclusion & Final Thoughts
🚀 Unleash the full power of AI image and video generation on the cloud! This comprehensive tutorial is your step-by-step guide to installing and configuring SwarmUI and ComfyUI on two of the most popular cloud GPU platforms: Massed Compute and RunPod.
Learn how to set up a powerful multi-GPU workflow to generate stunning, ultra-realistic images and videos at incredible speeds. We'll cover everything from deploying your first machine to downloading models, importing our exclusive presets, and running your first generations. Most importantly, we'll show you how to manage your instances to save money!
Whether you're new to cloud computing or looking to scale up your AI art projects, this guide has you covered.
💻 In this tutorial, you will learn how to:
Part 1: Massed Compute
Deploy a high-performance, multi-GPU machine.
Connect to your remote desktop using the ThinLinc client.
Install ComfyUI and the pre-installed SwarmUI from scratch.
Use the SwarmUI Model Downloader to get all the necessary models and bundles.
Configure SwarmUI backends to utilize multiple GPUs simultaneously for maximum speed.
Generate images and access your files from your local computer.
Properly terminate your machine to stop billing.
Part 2: RunPod
Deploy a multi-GPU pod using the correct PyTorch template.
Troubleshoot common connection issues.
Install both ComfyUI and SwarmUI in your RunPod workspace.
Set up SwarmUI backends for parallel processing on multiple GPUs.
Download models and presets for immediate use.
Understand the difference between stopping and terminating a pod to manage costs effectively.
Video Transcription
00:00:00 Greetings everyone. Today I am going to show you how to install SwarmUI and ComfyUI on
00:00:06 Massed Compute and RunPod and use all the presets we have, generate amazing images
00:00:12 or videos. If you remember, I had shown how to generate ultra-realistic images with SwarmUI by
00:00:20 using the Wan 2.2. We had published the tutorial for Windows like two days ago,
00:00:26 so please watch this tutorial to learn how to use it in details. Today I will show how to install
00:00:33 and use it on Massed Compute and RunPod, but it won't be as detailed as this one.
00:00:39 So first of all, let's download the ComfyUI installer. It is version 57 right now. You see
00:00:45 the latest zip file is here. Also, I recommend you to scroll down and read the news. You see,
00:00:52 I have compiled a new stage attention and it has fixed black outputs on RTX 3000 series.
00:01:00 Maybe more of the black outputs are fixed, I don't know yet, but this is an improvement.
00:01:06 Then download the SwarmUI installer. It is version 89 right now. We are updating,
00:01:13 adding more features. You can read all the news by scrolling down and see the
00:01:19 changelogs. Then move the zip files into any folder. So I moved them here. First of all,
00:01:25 I will extract the ComfyUI. Extract all, like this.
00:01:30 Then enter inside the folder and you will see that there is Massed Compute instructions .txt file.
00:01:35 Double click it. We have some links and tutorials here if you want to watch them later. Please use
00:01:40 this link to register. After registration, go to billing, set up some balance, then go to deploy.
00:01:48 In here, select your GPU. RTX Pro 6000 Blackwall GPU is the best, but L40S is also really good.
00:01:56 Let me show you their prices. So select creator, select the SECourses, apply direct coupon like
00:02:03 SECourses, verify. So RTX Pro 6000, 96 gigabyte GPU is 1.47 dollars per hour. Let's see the
00:02:12 L40S. Currently there is no L40S available, so I will use this one. Or there is only one GPU,
00:02:19 so let's select another one. For example, let's get H100. Okay, it is also a single
00:02:25 GPU left. So let's get this one. This is also a fast one. Okay, it is also single. Wow. Okay,
00:02:30 this one has two GPUs. Why I am choosing two GPUs? So I can show you how to utilize
00:02:35 both of the GPUs at the same time. Okay, click verify. You see 1.80 dollars for 2x A100. Deploy.
00:02:43 So this is important, we have selected image SECourses. Okay, now we need to wait for our
00:02:49 machine to initialize. Meanwhile, if you have not previously installed ThinLinc client,
00:02:55 go to ThinLinc client. You see from details, select the Windows installer,
00:03:00 or if you are using Mac or Linux select them. Open the installer, it will ask you,
00:03:05 click yes. Click next, accept, next, install. That's all default. Run ThinLinc client.
00:03:12 Important thing here is click options, go to local devices,
00:03:16 select clipboard synchronization and drives. Then click details. Then we are going to add
00:03:21 a folder. So select your folder from your PC, any folder you wish. This is my folder,
00:03:27 let me show you. You see, this is my folder. Then make the permission read and write to not
00:03:33 have any issues. Okay. Then okay. Okay, now we just wait for initialization.
00:03:39 Okay, the initialization of the machine has been completed. We are updating our image to
00:03:45 become available faster. So hopefully the machine initialization will become much faster. Then copy
00:03:52 this login URL and copy your password and copy your username. This is how we connect to our
00:03:59 machine. Click connect, continue. Wait for screen to start. Yes, it is starting. Then click start.
00:04:06 So this will connect to the remote machine, the Massed Compute desktop interface. This is running
00:04:13 on remote. I will copy my files into shared folder. You see ComfyUI and Swarm Downloader,
00:04:21 I will copy them. This is my shared folder. I will paste them here.
00:04:26 Then inside the ThinLinc client, it may take a while to load depending on your location,
00:04:32 your internet connection. Currently it is slow at me. Go to home, then scroll down and go to
00:04:38 ThinDrives. This is your shared folder with your computer. Whatever you put inside here,
00:04:45 whether on your PC or in Massed Compute, it will be synchronized. So you can download
00:04:50 your generated images and videos from here as well. So ComfyUI version 57 and SwarmUI model
00:04:57 downloader. Copy both of them or you can drag and drop into downloads. Or right click here
00:05:02 and paste. So you can use either drag and drop or copy and paste. Wait for synchronization to
00:05:08 be completed. You see, it shows that there is a synchronization here, copying files. You need
00:05:13 to wait this. Do not run anything inside shared folder, always copy into downloads.
00:05:18 So the ComfyUI has been copied. Right click and extract here. I will first install the ComfyUI.
00:05:24 You see there is Massed Compute instructions, double click it. Then select this install command,
00:05:29 right click, copy. Go back to folder, click this three dots icon. All of my installations
00:05:36 are same. Open in terminal. All of my applications installed same way. Right click and paste and hit
00:05:43 enter. This will start the ComfyUI installation. Now just wait it to be completed. This should be
00:05:48 fairly fast on Massed Compute. We are also working update of the image. As I said,
00:05:54 we will have the latest ComfyUI, latest SwarmUI and latest of everything. So the installations
00:05:59 will get faster, but it is already really fast. When you see the software updater,
00:06:04 just click cancel. You don't need to install anything. Just click cancel and follow the
00:06:09 CMD terminal from here. You see when I click it, it will show the terminal.
00:06:13 Okay, so the ComfyUI installation has been completed. Now we can move to the next
00:06:17 step, SwarmUI installation. However, it is not needed in our machine. So the ComfyUI installation
00:06:25 has been completed. Now it is time for SwarmUI, but SwarmUI has been installed already. So control
00:06:31 alt D, it will minimize everything. Then you see there is run SwarmUI update. Double click
00:06:38 this. This will update the SwarmUI to the latest version and start it. Just wait for it to start
00:06:44 at the beginning. So it is doing the update. It is started. Now you can close this. Okay,
00:06:49 it is already opened. I will just close it. I will first download models, then I will start
00:06:54 again. I will show you in a moment. So go back to downloads. Right click and extract the model
00:07:00 downloader. Enter inside the model downloader folder. You will see that Massed Compute model
00:07:06 download instructions .txt file. Copy this, right click and copy. Go back to your folder,
00:07:11 click this three dots, open in terminal and paste it. This will start model downloader
00:07:17 because we need to download models. If you put the models inside the Massed Compute image,
00:07:23 it takes too long to start. Therefore, we are removing all the models, but we will keep the
00:07:29 SwarmUI installed and few other applications installed. So the ComfyUI installation or update
00:07:34 will be faster. SwarmUI is already installed, as you have seen, it was just immediately updated.
00:07:41 But we are working on to make this machine start faster and more lightweight. So the application
00:07:48 is starting. In a moment, yes, started. It will be automatically opened like this.
00:07:53 So I am going to download SwarmUI bundles and image bundle. Let's see the download speed. It
00:07:59 should be blazing fast. Okay, 200 megabytes per second. Okay, around 300 megabytes per second,
00:08:07 very decent. You can download any model as well or all the bundles from here.
00:08:13 You can type the name like one, it will list you, or you can use URL downloader to download
00:08:18 from Civitai or anywhere you want. But it is downloading right now. If you want to learn more,
00:08:24 please watch this tutorial because I have explained it in this base Windows tutorial.
00:08:29 I am not going to repeat everything here again. So let's just wait for downloads to be completed.
00:08:35 Okay, once the model downloads have been completed, let's return back to desktop.
00:08:40 Control alt D, it will minimize everything. Then I am going to start the SwarmUI with
00:08:46 this one. You see run Cloudflare SwarmUI. Okay, let's just run it. It will start SwarmUI with a
00:08:53 Cloudflare link. Okay, it is started. You see here, open link. And copy this link like this
00:08:59 and open it in your browser so that it will work really fast. It will be running on the
00:09:06 Massed Compute machine, but I will be able to use it from my computer. So first of all,
00:09:11 we will add new backends. You see it is using our existing backend. We will update it later,
00:09:17 but using the installed ComfyUI is the best. So it is here inside ComfyUI. So click control L,
00:09:25 select it like this, control C, and paste it here. So click this edit icon, paste here,
00:09:31 and type main.py like this. Let me zoom out. Okay, I cannot zoom out. Maybe from here,
00:09:37 yes. You see this is the path wherever I have installed, it is ComfyUI installation and at
00:09:43 the end main.py. I am also going to add some extra arguments, --use-sage-attention. If you get black
00:09:52 outputs with your used model or preset, you can remove this. And I will make the OverQueue 0.
00:09:59 So this is for the first GPU. You can see that how many GPUs you have. We have two
00:10:04 GPUs right now. So let's go back to backends and add another Comfy self-starting. Copy this
00:10:10 path again and paste here. Copy the extra arguments. Make the GPU ID 1, make the
00:10:16 OverQueue 0 and save. So now when I generate multiple images or videos at the same time,
00:10:21 it will be distributed on each GPU. Go back to generate. Go to models and refresh. Yes, all the
00:10:28 models are here. Go to presets. We don't have the presets yet. However, we have preset here,
00:10:33 so I will right click and extract this into my own folder. So enter inside the folder and then
00:10:41 import preset, choose file. Go back wherever you have extracted it, select the preset and overwrite
00:10:50 existing ones if you want to overwrite, import. It is imported. Click refresh, sort by name.
00:10:57 So the rest is exactly same as in Windows tutorial, but I will make a demo for you and let's
00:11:04 also see the speed. First, we have to wait for backends to be loaded. When we go to the logs and
00:11:09 debug, we will see that it is loading. We should just wait. If it doesn't load for a long time,
00:11:15 you can just terminate all the terminals, quit all three terminals, then run the Cloudflare
00:11:21 stable SwarmUI again. It should fix the issue, most likely case. So let's see. Sometimes it may
00:11:29 take a while for Cloudflare URL to start. Yes, it is taking some time. You can also use from
00:11:36 the local URL it has from here. You see local URL. Let's go to logs. Okay. Do we have any
00:11:43 issues anywhere? Yeah, the backends loaded. Now we need to wait for Cloudflare. For some reason,
00:11:50 it didn't start. So if it doesn't start, you can restart again. Yeah, the Cloudflare didn't start,
00:11:56 so I will restart again. This can happen, unfortunately, so you need to try. Okay,
00:12:01 it gave us a new link. Okay, this one started. So let's open this in our browser.
00:12:07 Now the backend will load much quicker. Sometimes it gets stuck, so just restart.
00:12:13 Yes, it is started very quickly. Presets are here. As I said, the rest is exactly as in the Windows
00:12:19 tutorial. So let's make a demo. Quick tools, reset params, and realistic images. Cinematic image of a
00:12:29 fast car. Then you can choose your aspect ratio or resolution and generate. So the first generation
00:12:36 will be done on the first model. Generate again. The second one will be in the second model. Let's
00:12:41 generate 10 images. So it will generate 10 images. The generations will happen on both of the GPUs.
00:12:48 First of all, it is loading both of the GPUs. So let's go to logs to see. After load has been
00:12:54 completed, it will be much faster. Okay, the first generation started. Now let's just see them here.
00:13:02 Yes, you see both of the GPUs are generating images at the same time. So this way you can scale
00:13:08 your generation speed. You can rent as many as GPUs having machines, like eight GPUs. You can add
00:13:15 eight backends and generate eight images at the same time. Modern GPUs are much faster, like RTX
00:13:22 6000 Pro or RTX 4090, but A100 will be also fast. You see like 3 second IT. And we got the images.
00:13:32 So you can from more download, download them to your computer, or you can go to your ThinLinc
00:13:38 client, go to home, go to apps, go to stable SwarmUI. This is where it is installed. Go to
00:13:44 output and go to local. So these are where the images will be generated. So you can copy this
00:13:51 local, go to your ThinDrives, enter inside your shared folder, paste it. So once this is pasted
00:13:59 here, when you go back to your shared drive on your PC, you will see the generated images
00:14:05 are also copied here like this. This is another way of copying, mass copying. Either way works.
00:14:12 So once you are done with your generations, you need to turn off your machine. The other
00:14:16 presets are same. You can just quick tools, reset params to default and select them,
00:14:21 but you need to download their models. And downloading their models are easy. From the
00:14:26 model downloader, we have bundles for everything. So delete the research if you have and check out
00:14:31 the bundles. Image generation bundle, Qwen image core bundle, Wan 2.2 core bundle, Wan 2.1 core
00:14:38 bundle, so flux models core bundle. You can download them and use other presets as well.
00:14:44 And now I will turn off my machine to not spend any time. So click delete and delete. Once you
00:14:51 delete, it will be all gone. If you stop it, it will not stop your billing. So you need to
00:14:56 delete this. And it is deleted. That's it. Now I will start the RunPod part, so you can also
00:15:02 watch that part to learn how to use on RunPod as well. Thank you so much. Let's continue.
00:15:07 Open the RunPod instructions .txt file. You can watch the links here. You can read it. I recommend
00:15:13 that. Please register the RunPod from here. After registration, sign in if it doesn't automatically
00:15:19 sign in you. Then go to billing and add some credits to your account. Then go to pods. Then
00:15:26 you see there is options here. Click this deploy. I recommend to use secure cloud and in here,
00:15:34 I recommend to click additional filters and NVMe disk and make this 100 gigabytes to get a decent
00:15:42 pod. Most of the times the pods of the RunPod can be broken. I will show with 2x RTX 4090. Select
00:15:50 2x, so I can show you multiple backends and how to generate multiple images or videos at the same
00:15:56 time. Now this is confusing so many people. We are not going to use this template. Why? Because
00:16:03 in the RunPod instructions .txt file, it tells you to use this template. So always follow the RunPod
00:16:09 instructions .txt files for my applications. So click change template, select PyTorch 2.2. Okay,
00:16:17 2 GPUs edit because we are going to increase the volume disk size to like minimum 200 gigabytes.
00:16:24 If you are going to download more models, you need more and set overrides. Then click deploy on
00:16:28 demand. Just wait pod to be ready. This should be fairly fast because we are using official PyTorch
00:16:35 2.2 template. Yes. It takes few seconds. Then click Jupyter Lab. Okay, it is not ready yet,
00:16:41 so you need to try again and again until it becomes ready. If it doesn't become ready
00:16:46 like in one or two minutes, delete the pod and get a new one. Unfortunately, because RunPod is
00:16:51 extremely unpredictable. Okay, it didn't start. I am trying again. Yes, it still didn't start.
00:16:58 Try again. Okay, it doesn't start, so I will get a new pod meanwhile to increase my chances. Okay,
00:17:05 filters are remaining. Let's make this 2x, change template, 2.2 edit, 200 gigabytes, set overrides,
00:17:14 and deploy on demand. So whichever starts first, I will delete the other one. Okay, try again. No,
00:17:21 I cannot access yet. In these machines, I am doing Qwen image full fine tuning. I am researching it
00:17:29 right now. Okay, this one also failed. So I don't need this one at the moment. The test
00:17:34 has been completed. So stop pod and terminate. I don't need it. So I am deleting everything. Okay,
00:17:42 one more time I am going to test. So the Jupyter Lab and the Jupyter Lab. I hope
00:17:48 one of them starts. Yes, this one started. So you see this is this one and this one didn't
00:17:54 start. So I am going to delete that one, which is this one. So stop the pod and terminate the pod.
00:18:01 Then I am going to upload files. First of all, I am going to install the ComfyUI. So
00:18:07 upload the ComfyUI zip file into here and let's refresh. Okay, extract archive. By the way, you
00:18:16 need to wait for upload to be completed. It is not completed yet. So the extract would not work. Yes,
00:18:22 it didn't work. Okay, it is completed. Now right click and extract archive. Refresh. Yes. Then open
00:18:30 RunPod instructions .txt file. Select this install command, terminal, copy paste and that's it. You
00:18:37 are ready. You just need to wait for installation to be completed right now for ComfyUI.
00:18:44 The installation speed on RunPod 100 percentage depends on the pod you got. Sometimes it can take
00:18:50 one hour, sometimes it can take five minutes. Extremely undependable and unpredictable. On
00:18:58 Massed Compute, it is always fast, but on RunPod, it is up to your chances. Moreover, if you use the
00:19:04 permanent storage, it is way slower than getting a new pod as I did. But permanent storage advantage
00:19:11 is that you can start multiple machines on the same storage and it is always kept as it is.
00:19:18 This machine looks like a decent speed. It started installation sooner than I expect, but let's see.
00:19:24 Okay, so the ComfyUI installation has been completed. Now we will install SwarmUI.
00:19:30 Upload the SwarmUI zip file into your workspace. Wait for upload to be completed. It is uploading
00:19:36 right now. And what if if you want to just use the ComfyUI? You can use it. We have the instructions
00:19:43 all here. So the upload has been completed. You see SwarmUI model downloader, extract archive.
00:19:50 Then it will extract like this. Find the RunPod SwarmUI install instructions. Then copy this
00:19:57 entire string like this and open a new terminal. If you get this, just ignore. Ignore. This is due
00:20:04 to internet connection. Okay, now it's fixed. And hit enter. Now we will install SwarmUI instantly.
00:20:12 You will see that because we have installed the ComfyUI. Okay, just wait. It will give us a secure
00:20:20 Cloudflare link to connect SwarmUI. You can also use the RunPod proxy, but I don't recommend it
00:20:27 because it is very problematic and slow. Okay, we are still waiting. Yes. Now the SwarmUI started
00:20:34 and we got the Cloudflare URL here. Let me zoom in to show you. So this is the Cloudflare URL. Click
00:20:43 it and the SwarmUI interface will start like this. Agree, customize, important. You can choose any,
00:20:50 next. This one, next. This is none. We don't install ComfyUI local. Important, don't forget,
00:20:57 none, next. I don't want to download anything. Next, and yes, I am sure, and ready. You see?
00:21:04 Now we are going to add two backends because we have two GPUs. Click the ComfyUI self-starting.
00:21:10 Okay. It shows double, but it is actually one. Let's just refresh. Go to server backends,
00:21:15 yes. And add another one. And that's it. Now our ComfyUI is installed here.
00:21:22 Enter inside it. You will see that there is main.py. Right click and copy path. These
00:21:28 errors are not important, they will get fixed. Click this icon, copy paste it here,
00:21:33 but put a backslash to the beginning because all RunPod paths start with this backslash.
00:21:40 I'm going to use Sage Attention. If you get black output, remove this, but you
00:21:45 shouldn't. Then make this OverQueue 0 and save. We will do the same in the here. So copy paste,
00:21:53 copy paste. Make this GPU ID 1 and make this 0 and save. Now wait for backends to start. It is
00:22:00 starting. It should be fairly fast. Meanwhile waiting this to make it ready, you see we have
00:22:07 no models yet. So we are going to use the SwarmUI downloader inside workspace. Find the RunPod model
00:22:14 download instructions .txt file, open it, copy this part, copy, control C, open a new terminal,
00:22:22 paste it and hit enter. This will start the model downloader where you can download our
00:22:27 bundles, any single model, anything. I have explained everything in this tutorial video,
00:22:33 so watch it. I am not going to show everything again. That tutorial is mandatory. And click
00:22:38 this link and you will get the model downloader interface. It will automatically recognize our
00:22:44 model folder. Let's go to SwarmUI bundles and download the image generation models. But you
00:22:51 can download any bundle or any single model. Everything is possible. You can just search,
00:22:57 it will list you all the models. You can use the URL downloader, everything is fine. Let's
00:23:01 see the download speed. Yes, really good, like 600 megabytes per second. This is amazing speed
00:23:08 and our SwarmUI is still loading the backends. You can go to logs, debug and see what is happening,
00:23:14 but it should become available very soon. If this doesn't start for any reason, just stop your pod,
00:23:22 start again or restart and run the commands again, it should work. But I think it will work right
00:23:27 away. It is just downloading few models. And our models are getting downloaded. My downloader is
00:23:33 doing hash check. It is extremely robust. So when you watch this tutorial, you will see everything.
00:23:38 Okay, so the model downloads have been completed. Let's go back to our SwarmUI. Let's click refresh.
00:23:45 The models should appear. Yes, models appeared, as you can see. Then let's go to presets.
00:23:51 Currently we don't have any presets, so we need to import them. Import presets,
00:23:56 choose file. In your extracted folder, right click the Swarm model downloader and let's extract it.
00:24:04 I'm using the WinRAR, you can use anything. Extract files. Okay. Yes to all. Enter inside
00:24:10 the extracted folder and select the amazing Swarm preset latest version. You can click overwrite and
00:24:16 import and it will import all of our presets. Then click refresh and sort by name and ready.
00:24:24 So the rest is exactly as in the Windows tutorial part, but let's make a demo. So I will select the
00:24:31 Wan 2.2 generate realistic images. So reset params to default and direct apply. Photo of
00:24:37 a an old wise man. Let's generate 10 images so we can see the speed of both GPUs. First
00:24:44 it will load models to the GPU. This takes time on RunPod, so we need to wait. You can watch the
00:24:52 logs and what is happening. Moreover, it is still waiting for connecting to the server.
00:24:59 RunPod is always very slow compared to Massed Compute. Okay, we are still waiting. Yeah,
00:25:05 we are waiting. Let's click generate again. It should see our command on the logs, but I still
00:25:12 don't see. Let's copy this. Okay, it started. You see? 20 current generations because we clicked
00:25:19 double times the generate, but it is slow. You just need to wait. Now in the debug, yes,
00:25:25 now it's starting to load models. Always patiently wait, it gets faster once the models loaded.
00:25:33 Okay, so the generations started. We can see the previews. For example, this is the image. The rest
00:25:40 is exactly as in the Windows tutorial part. You can of course always apply the image upscale which
00:25:47 improves the quality significantly. To download these images, you can click more and download or
00:25:52 go back to your workspace inside SwarmUI, inside output. You can download from here. Right click
00:25:59 and download as an archive. It will download all the generated images and videos. If you want to
00:26:05 use these presets, you just need to download their models. In our model downloader, we have bundles.
00:26:10 You see Swarm bundles, image generation bundle, Qwen image core bundle, Wan 2.2 core bundle. For
00:26:16 generating videos, I recommend this. Wan 2.1 core bundle, flux models core bundle. Once you download
00:26:22 those bundles, these presets will become available to use to you. And that's it. I hope you have
00:26:28 enjoyed. Don't forget to turn off your machine once you are done. It's inside pods. So here the
00:26:37 machine, I can turn it off from stop pod and I can start again. This is one of the advantage of
00:26:43 RunPod. All the data will remain. After starting again, I just need to start the SwarmUI same as I
00:26:50 have installed. It will be instant and all will be ready. However, you see this machine will
00:26:55 use 6 cents per hour until I terminate it. When I click the terminate pod, it will be gone forever.
00:27:04 For example, this pod is running on my storage from here. So in this pod, there is no stop pod.
00:27:10 I can terminate it, but all the data will remain in my storage. It's exactly same logic. You just
00:27:16 make a new network volume and the rest is same. So hopefully see you later. Thank you so much.
