Master Local AI Art & Video Generation with SwarmUI (ComfyUI Backend): The Ultimate 2025 Tutorial

[FurkanGozukara]

Master Local AI Art & Video Generation with SwarmUI (ComfyUI Backend): The Ultimate 2025 Tutorial

Full tutorial: https://www.youtube.com/watch?v=fTzlQ0tjxj0

If you want to generate the very best AI videos and images on your Windows computer locally this is the tutorial that you were looking for. Literally 1-click to install most powerful and advanced generative AI interface SwarmUI (with Flash Attention, Sage Attention, Triton, DeepSpeed, xFormers, RTX 5000 series perfect compatibility) and download the very best AI image and video generation models with ultra advanced model downloader Gradio app. SwarmUI utilizes the famous and most powerful, advanced, performant and optimized ComfyUI as a backend. So SwarmUI is the ultimate generative AI tool at the moment with vast amount of features and constant updates.

🔗Follow below link to download the zip file that contains SwarmUI installer and AI models downloader Gradio App - the one used in the tutorial ⤵️

▶️ https://www.patreon.com/posts/SwarmUI-Installer-AI-Videos-Downloader-114517862

🔗Follow below link to download the zip file that contains ComfyUI 1-click installer that has all the Flash Attention, Sage Attention, xFormers, Triton, DeepSpeed, RTX 5000 series support ⤵️

▶️ https://www.patreon.com/posts/Advanced-ComfyUI-1-Click-Installer-105023709

🔗 Python, Git, CUDA, C++, FFMPEG, MSVC installation tutorial - needed for ComfyUI ⤵️

▶️ https://youtu.be/DrhUHnYfwC0

🔗 SECourses Official Discord 10500+ Members ⤵️

▶️ https://discord.com/servers/software-engineering-courses-secourses-772774097734074388

🔗 Stable Diffusion, FLUX, Generative AI Tutorials and Resources GitHub ⤵️

▶️ https://github.com/FurkanGozukara/Stable-Diffusion

🔗 SECourses Official Reddit - Stay Subscribed To Learn All The News and More ⤵️

▶️ https://www.reddit.com/r/SECourses/

This tutorial will show you how to install ComfyUI properly into a VENV with 1-click our installers, then install SwarmUI instantly and then how to give propoerly installed ComfyUI as a backend to the SwarmUI and generate AI images and videos with maximum speed and quality. Moreover, this tutorial will introduce you ultimate AI model downloader Gradio app being developed by us.

00:00:00 Introduction: Fastest Local Image & Video Generation with SwarmUI on Windows

00:00:38 Leveraging ComfyUI Performance & Sage Attention Speed Boost in SwarmUI

00:01:07 Introducing the Custom Unified Model Downloader Application

00:01:50 Overview of Supported Models (SD, Video, Text Encoders, GGUF, VAE etc.)

00:02:50 Why Manually Installed ComfyUI Backend is Better for SwarmUI Performance

00:03:07 Live Demo Preview: High Dream GGUF Image Generation on Dual GPUs (RTX 5090 + 3090 Ti)

00:03:52 Tutorial Start: Overview of Installers (SwarmUI, Model Downloader, ComfyUI)

00:04:20 Step 1: Downloading the SwarmUI Model Downloader Installer Zip

00:04:33 Step 2: Downloading the Manual ComfyUI Backend Installer

00:04:52 Step 3: Extracting ComfyUI Installer for Backend Setup

00:05:05 Step 4: Running ComfyUI Backend Installation (windows_install.bat)

00:05:33 Configuring ComfyUI Install: Python Version & Attention Mechanisms (Flash/Sage)

00:06:18 Step 5: Starting SwarmUI Frontend Installation (Parallel to ComfyUI)

00:06:40 Configuring SwarmUI Installation Options (Theme, Custom Settings)

00:06:54 CRUCIAL STEP: Selecting "None" for ComfyUI Backend during SwarmUI Setup

00:07:21 Step 6: Adding the Manually Installed ComfyUI Backend in SwarmUI Settings

00:07:43 Configuring Backend Path (main.py) & Enabling Sage Attention (--use-sage-attention)

00:08:28 Monitoring Backend Dependency Installation & VENV Explanation

00:09:17 Step 7: Adding a Second ComfyUI Backend for Multi-GPU Usage (GPU ID 1)

00:10:00 Step 8: Launching & Setting Up the Unified Model Downloader App

00:10:50 Using the Model Downloader: Downloading Flux Models Bundle Example

00:11:14 Understanding Downloader Speed (HF Transfer) & Progress Monitoring

00:11:50 Downloader Features: Queue System & No Re-Downloads (Wan 2.1 Example)

00:12:20 Accessing 'How-to-Use' Links for Models within the Downloader

00:13:22 Step 9: Starting SwarmUI & Refreshing Models After Downloading

00:14:08 First-Time GGUF Model Load: Installing GGUF Support Extension in SwarmUI

00:14:32 Manually Setting Model Metadata (Architecture) if Needed (HiDream Example)

00:14:48 Detailed GGUF Support Installation Walkthrough & Backend Restart

00:15:51 Ignoring C Link Fatal Error on Windows

00:16:23 Usage Demo: Generating Image with Custom Flux DreamBooth Model

00:16:41 Configuring Generation Parameters (Prompt, Steps, Sampler, CFG, Guidance)

00:17:07 Initiating Generation & Monitoring Model Loading on GPUs

00:17:50 Troubleshooting Face Inpainting & Downloading YOLO Models via Downloader

00:18:31 Advanced Feature Demo: Using the Refine Upscale Option

00:18:43 Configuring Upscale Settings (Refiner %, Upscaler Model)

00:19:19 Running Generation with Upscaling & Performance Considerations

00:19:40 Comparing Results: Original vs. High-Quality Upscaled Image Detail

00:19:52 Fine-tuning Face Inpainting: Adjusting Mask Parameters (Blur/Grow)

00:20:23 Conclusion & Where to Find Links and Ask Questions

Music provided by NoCopyrightSounds :

Music Credits : https://gist.github.com/FurkanGozukara/8fa15b67370a53af6394da239d2fce8e

Video Transcription

• 00:00:00 Greetings everyone. Today I am going to show you  the latest method, the most accurate, and the most  

• 00:00:07 performant method of generating images and videos  on your Windows local computer with the one-click  

• 00:00:15 installation and setup. I am going to show you  how to use SwarmUI and how to download models  

• 00:00:22 and start using them immediately on your Windows  computer. This will be an epic tutorial that will  

• 00:00:30 make your local image and video generation many  times easier and with the maximum optimization.  

• 00:00:38 Maybe you heard of the ComfyUI. The ComfyUI is  so hard to use, but we can still leverage its  

• 00:00:46 performance, its features with the SwarmUI. And  my installation even supports out-of-the-box Sage  

• 00:00:53 Attention. Maybe you also heard of it. It is the  fastest attention that makes your generations way  

• 00:01:00 faster for both image and video generation.  And I will introduce you the application  

• 00:01:07 that I have developed for downloading models  with the maximum speed and maximum easiness. 

• 00:01:13 So far we have two model bundles: Flux model  bundle and High Dream L1 model bundle. With just  

• 00:01:21 clicking these buttons, it will download all the  necessary files into your SwarmUI installation.  

• 00:01:27 Moreover, we have all kind of models with  one-click to download. So you will not  

• 00:01:33 spend time for finding models, downloading  them into the accurate folders. Moreover,  

• 00:01:38 I also have how-to-use links. So when we go  to these links, we will learn how to use this  

• 00:01:45 application, this model in the SwarmUI itself. You see, we support High Dream, we support  

• 00:01:50 Stable Diffusion 1.5 models, we support Stable  Diffusion XL models, we support Stable Diffusion  

• 00:01:56 3.5 large models. We support YOLO Face inpainting  models. We support text encoder models, T5-XXL,  

• 00:02:04 UMT T5-XXL, Clip models, LLM text encoder. We  support video generation models: Wan 2.1, Hunyuan,  

• 00:02:14 Fast Hunyuan models. We support SkyReels. And you  see that we support GGUF and also BF16 and FP16,  

• 00:02:23 FP8 versions. We support Genmo Mochi 1, Lightricks LTX video, VAE model. 

• 00:02:30 And this application will get better and better  as you make requests to me. So you can tell me  

• 00:02:36 that add to this model, add to that model,  and hopefully I will update it. In the very  

• 00:02:40 top of the application, you can also search  and filter with the model name. So it is so,  

• 00:02:47 so easy to use and so easy to install. Using SwarmUI with a manually installed ComfyUI  

• 00:02:54 backend is working better than the pre-compiled  ComfyUI backend. So today I will also show you  

• 00:03:01 how to use properly installed ComfyUI backend in  the SwarmUI for maximum performance. By the way,  

• 00:03:07 these images that you are seeing right now are  being generated with High Dream L1 Dev GGUF  

• 00:03:14 Q8 model. GGUF Q8 is almost same quality as the  highest quality BF16 model itself. Currently I am  

• 00:03:23 generating with my dual GPUs. I have RTX 5090 and  I have RTX 3090 Ti. So you see both of the GPUs  

• 00:03:32 are running almost at full power. And also I am  recording a video right now, so you see my PC is  

• 00:03:39 under heavy load. However, SwarmUI with ComfyUI  backend is still able to handle generations  

• 00:03:47 really, really fast. They are getting generated  right now live as you are seeing right now. 

• 00:03:52 So, let's begin the tutorial and how  to install and use these applications. 

• 00:03:57 So, we have SwarmUI one-click installer with  the SwarmUI Unified Model Downloader. But this  

• 00:04:04 Unified Model Downloader can also be used  for ComfyUI or for other applications that  

• 00:04:10 you use as well. It supports custom folder  paths and also ComfyUI installer. The links  

• 00:04:16 of these installers will be in the description  of the video, so check them out. First of all,  

• 00:04:20 let's begin with downloading the latest SwarmUI  Model Downloader zip file. So, latest zip file  

• 00:04:26 is SwarmUI Model Downloader version 29. I have  just updated it to the latest version. It is  

• 00:04:33 downloaded. Let's also download the manual  installer of ComfyUI. By manual, I mean that  

• 00:04:39 we are not using pre-compiled. We are going to  install with our installers. Move the downloaded  

• 00:04:44 files into any of your drives. I will use my E  drive. First of all, let's extract the ComfyUI  

• 00:04:52 installation. This will be used as a backend. And  why not use the pre-compiled backend? Because this  

• 00:04:59 is way better. When you use the pre-compiled  backend, it is not very up-to-date, it causes  

• 00:05:05 problems when installing new libraries. So I  will use the windows_install.bat file. This is  

• 00:05:12 by default supporting xFormers, Flash Attention,  Sage Attention, Triton, whatever you need. And it  

• 00:05:19 works on all GPUs including RTX 5000 series. I recommend to use Python 3.10. You need to  

• 00:05:27 have Python installed on your computer. So if you  are using another version, you can select it. I  

• 00:05:33 really recommend Python 3.10. So I will select  option 1. So you can choose Flash Attention,  

• 00:05:39 Sage Attention, or both attentions. Let's  choose the both of the attentions. And it  

• 00:05:44 will also install the xFormers as well. This is  going to install with latest official Torch 2.7.  

• 00:05:51 Don't worry, it will work on all GPUs and also  pre-compiled Flash Attention, Sage Attention,  

• 00:05:58 xFormers. I have compiled myself xFormers  and Flash Attention, and the Sage Attention  

• 00:06:03 and Triton is from our legendary woct0rdho. Maybe  you heard this guy. This guy is the sole developer  

• 00:06:13 of Triton and Sage Attention for Windows. So, meanwhile this is getting installed,  

• 00:06:18 we can also install SwarmUI. We don't need to  wait. So let's just extract the SwarmUI Model  

• 00:06:23 Downloader. Extract all, and here. So let's  begin with installing SwarmUI. You see we have  

• 00:06:29 windows_install_swarmui.bat file. This is from  the official repository, nothing is special here.  

• 00:06:34 It will just start the installation like this. The  options that you are going to choose is important.  

• 00:06:40 We are going to choose Agree, Customize Settings,  Yes. Modern Dark or whatever the you wish,  

• 00:06:48 just yourself. And this is the most crucial  part. We are not going to use ComfyUI Local.  

• 00:06:54 This is pre-installed, pre-compiled ComfyUI. I  do not recommend this. So let's go with None,  

• 00:07:00 and Next. You can download, but I don't recommend.  Our downloader is faster. Next, and Yes,  

• 00:07:06 I am sure to install and installed. That's it. So the SwarmUI has been installed. Let's look  

• 00:07:11 at the ComfyUI installation. It is still getting  installed as you are seeing right now. All right.  

• 00:07:16 The ComfyUI has been installed. You see, press  any key to continue. And the SwarmUI has been  

• 00:07:21 started. Now what we need to do is adding backend.  So click Server, click Backends, and get to this  

• 00:07:29 page and click ComfyUI Self-Starting. Click OK.  And it shows two backends. This is a bug. Let's  

• 00:07:36 refresh and it is fixed. So let's check this,  click this green icon for editing. We need to  

• 00:07:43 give path of the ComfyUI here. So let's return  back to our ComfyUI installation. It was here.  

• 00:07:49 Enter inside the ComfyUI and you see there is  main.py file here. So let's copy like this, and  

• 00:07:56 /main.py file. This is the backend. If you want  to use Sage Attention, --use-sage-attention like  

• 00:08:06 this. Sage Attention works only RTX 3000 and newer  GPUs. So like 4080 or 5060. So 3000, 4000 and 5000  

• 00:08:16 series are supporting. But don't worry, if your  GPU is not supporting, it will use either xFormers  

• 00:08:23 or it will use the Torch attention itself. So wait for backends to be loaded. When we go  

• 00:08:28 to the Logs and Debug, it will show us what  is happening right now. So it is installing  

• 00:08:34 the missing libraries into our manually installed  ComfyUI. And our ComfyUI has been installed with a  

• 00:08:42 virtual environment. So this installation will not  break any other applications or installations that  

• 00:08:48 I have. This is 100 percentage isolated to here.  So it is 100 percentage secure. With this way,  

• 00:08:55 you can have as many as different ComfyUI  installations and we will use the model  

• 00:09:01 folder of the SwarmUI itself. Or you can  give any model folders. How? When you go  

• 00:09:06 to the Server Configuration, you can set  your model root and other model paths  

• 00:09:11 from here. Okay, so the installation has  been completed. I have my backend ready. 

• 00:09:17 Since I have also another GPU, let me show you.  So open a CMD, type pip install nvitop, then type  

• 00:09:25 nvitop and you will get to this page. By the way,  yes. So let's start again. nvitop, and here. So  

• 00:09:32 I have RTX 5090 and I have RTX 3090 Ti. So I can  also utilize my second GPU. To do that, let's copy  

• 00:09:41 this, click ComfyUI Self-Starting, OK, and paste  it here. So copy this, paste here and give the GPU  

• 00:09:48 ID 1 and Save. So this will work on the second GPU  and this will work on the first GPU. And we are  

• 00:09:55 ready. Our SwarmUI has been started. However,  currently we do not have any of the models. 

• 00:10:00 So, to download the models, let's return  back to our SwarmUI folder where we have  

• 00:10:06 extracted the files. You see here. Click  windows_start_download_models_app.bat file. More  

• 00:10:11 info, Run anyway. It will generate a VENV virtual  environment folder here, install the necessary  

• 00:10:17 libraries, which is not much, just Gradio and  some other libraries, then it will start the  

• 00:10:23 application. Let's just wait. For this to work,  you also need to have Python installed. So if you  

• 00:10:29 don't know how to install Python and the other  libraries, I have an excellent tutorial video.  

• 00:10:34 The link will be in the ComfyUI installation post  as you are seeing right now, Windows Requirements.  

• 00:10:40 So you can watch this tutorial and learn it  with full details and step-by-step guidance. 

• 00:10:45 So the Model Downloader started. This is the Model  Downloader interface. Whichever the model you want  

• 00:10:50 to use, you can search here and download.  This is an extremely advanced UI. And also,  

• 00:10:56 if a model is missing here, just message me from  Patreon and hopefully I will add it as soon as  

• 00:11:02 possible. So for demonstration, we can download  Flux Models bundle. It will download all these  

• 00:11:09 models. So let's click Download Flux Models  bundle. And you can watch the status here. So  

• 00:11:14 it will utilize my entire network right now. Why?  Because we are utilizing enable HF transfer. This  

• 00:11:21 is a special download method and it will speed up  your downloads to your maximum network speed. Let  

• 00:11:28 me show you from Task Manager, Performance, and  you see I am downloading with 500 megabits per  

• 00:11:34 second, like 600 megabits per second. I am also  uploading models into more optimized Hugging  

• 00:11:42 Face XET repositories so that it will be even faster  hopefully in future without any errors. Moreover,  

• 00:11:50 this downloader has queue system and if a file  has been downloaded previously, it will not  

• 00:11:57 re-download it. So this is an amazing feature.  For example, I can also queue video generation  

• 00:12:03 models download, like Wan 2.1 models download. So  I can click this and it will queue that. Let's  

• 00:12:09 see. So you see the queue size is now 7 and it  shows that queued Wan 2.1 text-to-video model. 

• 00:12:16 Moreover, if you don't know how to  use these models inside SwarmUI,  

• 00:12:20 just click these links and read them. For example,  this is Wan 2.1 parameters in SwarmUI. You can read  

• 00:12:28 here and learn it. Hopefully I will make different  tutorial videos, so stay subscribed to our channel  

• 00:12:33 to not miss any of them. This application works  on RunPod and also on Massed Compute as well. The  

• 00:12:40 instructions are all in the zip file: Massed  Compute instructions, RunPod instructions.  

• 00:12:45 Hopefully I will make a separate video for them to  not make this video very long. So stay subscribed.  

• 00:12:52 And this will download all of the models with  the maximum speed, then you will be able to  

• 00:12:57 use models inside your SwarmUI installation. The  models will be automatically downloaded into the  

• 00:13:03 accurate folders, so you don't need to spend  time with them. It will show you everything,  

• 00:13:09 every information on the screen and also on the  terminal, on the CMD window. If you have any  

• 00:13:15 questions, just ask me and hopefully I will answer  them. And how to use these models are so simple. 

• 00:13:22 Let me start my other installation which I  have been showing you in the beginning of  

• 00:13:26 the video. So for starting the SwarmUI after  installation and models have been downloaded,  

• 00:13:32 just click the Windows_start_swarmui.bat file.  SwarmUI now automatically updates, so there is no  

• 00:13:39 update button anymore. Then once the models have  been downloaded, go to Models, click Refresh and  

• 00:13:46 select the model that you want. One of the amazing  feature of our downloader is that we download the  

• 00:13:52 highest quality T5-XXL and we give you option to  download highest quality models as well, like Flux  

• 00:14:00 Dev FP16 model or you can also download Flux Dev  GGUF model. When you first time try to run a GGUF  

• 00:14:08 model, it will ask you to install GGUF extension.  It is important. And also in some models, it may  

• 00:14:16 require you to set the model type. For example,  this is automatically recognized High Dream L1 Dev  

• 00:14:23 GGUF Q8 model. So when I click this hamburger menu  and say Edit Metadata, you can set it from here,  

• 00:14:32 like High Dream architecture. It usually recognize  all of the models, but if it doesn't recognize,  

• 00:14:38 select the architecture from here  and save it, then it will work. 

• 00:14:42 Let me also show you the GGUF installation so you  will not have any issues with that. To do that,  

• 00:14:48 I will just copy my previous model to not wait  it downloading. So the models are saved inside  

• 00:14:55 diffusion models usually. So this model is saved  inside the diffusion models. So let's copy this  

• 00:15:02 and go back to our new installation. Just copy  paste it. You see our download is continuing. It  

• 00:15:08 downloaded the Flux Dev model, now downloading  Flux Fill Dev model. And return back to our new  

• 00:15:13 installation. It is here and the models are here.  So let's refresh. Yes, arrived. When I click Load  

• 00:15:20 Model, it will ask me Install GGUF support. You  have to install this to be able to run GGUF. So  

• 00:15:27 Install, OK. You need to wait backends to be  loaded, then it says Installed. Please wait  

• 00:15:34 while backends restarting. If it doesn't work,  you may need to restart SwarmUI, but it usually  

• 00:15:39 works. Just follow the logs from here, Debug  menu, and you will see that it is installing  

• 00:15:44 necessary libraries and starting. And after it has  been started, it will load model into the VRAM,  

• 00:15:51 into my GPU. You will also get this error. You see  test C link fatal error. This is a Linux related  

• 00:15:59 error. Since we are running on Windows, you  can just ignore this. This happens in my all  

• 00:16:05 applications, but you can safely ignore it. And  the backend has been loaded and it is ready to use  

• 00:16:11 right now with the GGUF support as well. You can  also install TeaCache and other stuff, but hopefully  

• 00:16:17 I will make different separate videos for them. So this is how you can begin using the very best  

• 00:16:23 models in your local computer after installing  and downloading them. For example, this is a  

• 00:16:29 DreamBooth model that I have trained myself on.  And let's generate an image as a demonstration  

• 00:16:35 before we end the tutorial. So we have an amazing  Kohya configuration. It is also on Patreon. Let's  

• 00:16:41 go to the Test Prompts. Let's select a Realism  prompt. For example, this one. This is a model  

• 00:16:46 that has been generated on me. Since this is a  Flux model, let's make the 40 steps sampling,  

• 00:16:52 iPNDM, my favorite sampling. CFG is 1. This is  for Flux and also High Dream is also same. Flux  

• 00:16:59 Guidance Scale 4. And that's it. Let's generate  an image. First, it will load it into the my  

• 00:17:07 RTX 5090, then it will generate the image.  Okay. Oh, we did set it as 100 images. Okay,  

• 00:17:13 let's see. And we have the two backends here, so  it will use both of them. All right. It should  

• 00:17:19 start loading in a moment. You can always watch  the logs from here. Let's go to Debug and let's  

• 00:17:26 see what is happening. Okay, it says that  it is waiting for model to be loaded. Yes,  

• 00:17:31 now it is loading. We can watch the status from  here. Yes, it is loading now first into the RAM,  

• 00:17:37 then it will move them into the GPUs. Okay, the  first GPU is warming up, loading the model. Yes,  

• 00:17:44 the generation has been started. We can see that.  Okay, let's see the image is getting generated. 

• 00:17:50 Then it will inpaint the face. I have been getting  asked of, the inpainting model is not working.  

• 00:17:56 Because your installation were not accurate. So  install as I have installed in this tutorial,  

• 00:18:01 then download the face models. For example, you  can... let's refresh this page. It shouldn't  

• 00:18:07 impact the download. You see it is returned  back. Go to the Other Models, and you see  

• 00:18:12 we have Auto YOLO Masking models. Then click here  and just download. It will download the best face  

• 00:18:19 automatic recognition, segmentation models.  And you see, it has generated the image. Oh,  

• 00:18:25 there is one more thing which I need to show you.  Let's cancel all sessions. Which is upscaling.  

• 00:18:31 This was also getting asked of me. What is the  best upscaling? Best upscaling is like this. Let  

• 00:18:38 me show you. So go to the Refine Upscale, enable  it first. If you don't see Advanced options, click  

• 00:18:43 here to see the Advanced options. Set the Refiner  percentage 60 percentage. Let's click Reuse  

• 00:18:49 Parameters to upscale this image. Enable Upscale,  Refiner 60 percentage, and Refiner Upscaler 2x.  

• 00:18:57 And this is super important. I use 4x RealWebPhoto  version 4. This is automatically downloaded with  

• 00:19:03 our downloader. Where? You see we have Other  Models here. When you click the Other Models  

• 00:19:08 and when you go to the Image Upscaling models,  you can download all of the image upscaling  

• 00:19:13 models by clicking here and it will automatically  download them for you into the accurate folder. 

• 00:19:19 Now first it will generate this image, then it  will upscale it. Upscaling is really heavy because  

• 00:19:26 from 1 megapixel to 4 megapixel we are upscaling  with Refiner Upscale 2x. But this is really,  

• 00:19:34 really next level. It really improves the quality.  I will show you in a moment. Okay, the upscale  

• 00:19:40 has been completed. Okay, now watch carefully.  Original image, upscaled image. Original image,  

• 00:19:46 upscaled image. You see the difference is huge.  And probably I don't even need to face inpaint  

• 00:19:52 this. You see the face inpainting has some  noticeable area here because it is black. So  

• 00:19:58 to remove this noticeable area there, you need to  play the parameters here like Segment Mask Blur,  

• 00:20:04 Segment Mask Grow, the other things to remove  it. I don't know the exact parameter right  

• 00:20:08 now. I don't remember, but you can pretty much  remove this as well. It is barely noticeable,  

• 00:20:13 but and also it can be really easily fixed  it. But from this quality into this quality.  

• 00:20:19 And this is amazing. This is next level. Thank you so much, guys. You can always  

• 00:20:23 message me from Patreon. You can also reply  to this video. Hopefully see you later.  

• 00:20:27 The links will be in the description of  the video and also in the pinned comment.