How I Make VIRAL Fitness Videos With AI | 100% FREE & Unlimited

0:00The faceless fitness niche is honestly one of the fastest growing and most profitable niches on YouTube right now. Some videos in the space are getting RPMs as high as $25 per 1,000 views. And just to give you some perspective, even if we lowball it and say a channel is making around a $10 RPM, one video hitting 1 million views could still

0:19bring in about $10,000. Now, lately, I've been coming across a bunch of fitness channels that are doing insanely well, and some of them are pulling in millions of views in just a couple of months. So, I went on YouTube to see if anybody had actually broken this niche down properly, but honestly, I couldn't really find a good tutorial that walked through the entire process step-by-step. So, in this course, I'm going to walk you through exactly how I launch my faceless fitness channels step-by-step. We're going to cover the whole process: creating the channel, branding it, writing scripts, generating visuals, making high-quality thumbnails, and then editing everything together

0:56into a finished video. Now, the crazy thing is, this exact method works for replicating basically any channel in pretty much any niche. And the best part? You can do this entire process from literally any device, even your phone. You don't need to install any software or anything like that. And to make it even better, we're going to be using 100% free AI tools for the entire

1:18workflow. In fact, check out this cool video I made in like 15 minutes just by following this free method. Most people think you have to choose between being a functional ninja or a bloated tank. They'll tell you body weight is just for warm-ups, or that the gym is the only place to build real size. Here's the truth: both sides are wrong. If you want to look like an athlete and move like one, you need to understand how muscle

1:42actually grows. Your body doesn't have a GPS for the gym. It only understands tension. [music] Whether that tension comes from a heavy barbell or a gravity-defying pull-up doesn't matter to your biceps. It's all about leverage and load. Let me explain. Most gym bros are just moving weight from A to B. Most calisthenics fans are

2:01just doing endless reps of easy moves. Both are wasting time. Stop picking sides. All right. Now, the very first thing you need is basically the brain of your entire operation. We need a chatbot that's going to help automate this whole process for us. Normally, I'd go with Claude, but since we're keeping this completely free, for this course, we're going to use Google Gemini instead. It's totally free. All you've got to do is sign in with your Google account, and you'll get an interface that looks something like this. Now, once you've got your chatbot ready, the next thing

2:28you need is the fitness master prompt. You should be able to find it either inside my completely free school community or through the link in the description. Once you've got access to it, simply copy the entire prompt, paste it into Gemini, and send it over. After you do that, Gemini is going to ask you for the PDF source file. Now, I've

2:44actually already set this up for you. So, if you go back to the fitness master prompt and scroll all the way to the top, you should see something called the fitness source PDF. Just click on that and download the file to your computer. Now, this file basically contains everything Gemini needs to understand the kind of content we're trying to make. Transcripts, reference channels, background music styles, animation

3:04styles, pretty much the whole blueprint. All right. Once you've downloaded it, head back into Gemini, upload the source file from your computer, and just type something like source file attached, then hit send. Now, when you do that, Gemini is going to go through the entire file, and it's a pretty long one, so give it a second to read through everything and understand the structure of what we're building. And once it's done absorbing all that information, all you've got to do is type go. Now, once you do this, Gemini starts giving you character creation prompts for your main character. And the reason this is important is because if we look at the

3:37reference channels in this niche, we can see they all use a consistent character throughout their branding. So, basically, their channel has a face to it. And in this niche, you need the same thing. You need to create a recognizable face for your channel. Something that shows up on your banner, your thumbnails, and even in your visuals. Now, Gemini is going to suggest around four different character variants you can work with, and honestly, you can pick any one of them. And if you don't like any of them, you can simply ask Gemini to generate more. You can specify the type of character, the colors you want, or any style you're going for. For

4:08me, I think this one looks really good. I like variant four, which is the green character. So, I'll just say I like variant four and send it. Now, after that, Gemini is going to suggest a couple of tools you can use to actually bring this character to life. And what we need here is a free and unlimited image creation platform. So, after doing some research, I found this platform called DeGen. So, you're going to search for DeGen AI and just sign in with your Google account. Now, once you're inside DeGen, you'll see it gives you access to a bunch of different models. And right now, I'm on the free plan, which comes

4:39with like 300 credits. But the cool thing is, if you're just doing image creation, they don't really eat up your credits the same way most tools do. So, once you're signed up, just click on the image creation tab, and it'll take you to a new interface that looks like this. Now, down here, you can pick from a bunch of models, but they clearly label which ones are free and which ones use credits. Right now, Flux 2 is basically the only fully free advanced model. You still have Nano Banana and Nano Banana Pro, but those ones will cost credits when you generate images. The good thing

5:09though is the credits refresh every day. So, for this tutorial, we're just going to stick with Flux since it's completely free. Set your aspect ratio to 16 by 9, then go back to Gemini. Scroll up to prompt variant four, the character we chose, copy it, paste it into the prompt box, and send it. Now, after a few seconds, you're going to get an output like this. And honestly, the characters look really good. Everything is pretty solid. And if you don't like what you see, you can always just go back to your

5:34chatbot and ask for another variation. But for me, this one looks pretty good. So, at this point, I'm simply going to download this character and save it to my computer. Now, just keep in mind, whatever character you pick here becomes the face of your entire brand. So, make sure you choose something you can comfortably use across all your videos long-term. Once we've locked in our character, I'm going to head back to Gemini and type next. Now, when you do that, Gemini starts generating branding suggestions for you. It'll give you around 10 different brand name ideas based on the character you picked. It also gives you a channel description, a

6:05logo prompt you can use to generate your logo, and even banner prompts as well. So, we're going to use all of those details later to fully brand the channel. But for now, we'll come back to that after we finish the video. Right now, let's just move on to the next step. Once you type next, Gemini will then start suggesting about 10 different video ideas for you to choose from. And you don't have to stick to these ideas if you don't want to. The whole point is just to help you generate content ideas consistently, so you never run out. Now, for me, I think I'm going to go with

6:33idea number three, and I'll send that. Then, Gemini is going to ask you how long you want your script to be. Now, since this is a tutorial, I'm going to set mine to 1 minute long. But if you're planning on making longer-form content for YouTube, yours should ideally be around 10 to 15 minutes. Because whatever duration you give Gemini is going to directly affect how long your script turns out. For example, I asked for a 1-minute script, and Gemini gave me something that was about 162 words, which is actually pretty solid and exactly what we want. Now that we've got our script, the next step is to turn

7:07this script into a voice-over file. Normally, I'd recommend using ElevenLabs because they have really high-quality voices, and they also give you around 10,000 free But since we're sticking with a completely free AI tool workflow, we're going to use a platform called Noise AI instead. So, just open your browser, search for Noise AI, and then sign in with your Google account. Now, Noise AI actually gives you about 2,000 credits every single day to generate voice-overs, which is a pretty big advantage. So, if you're sticking with the free option, this is going to be your go-to. Of course, ElevenLabs still gives you more premium quality outputs, but since we're keeping this workflow

7:41completely free, we're not going to break that streak. So, all you really need to do is click on the voices option, and that'll bring up a list of different voices you can choose from. For a channel like this, you're going to want a male voice. So, just scroll through, find a male voice that fits the

7:55style you're going for, and select it. Then, the next step is simple. Go back to Gemini, copy a chunk of your script, paste it into Noise AI's Chrome box, and then just click speak it to generate your voice over. Now, once your voice over is ready, all you've got to do is click the download option, and it'll save straight to your computer. Then, you just go back to Gemini, copy the second half of your script, paste it into Noise AI, and repeat the exact same process until you've covered the entire script. Now, once that's done, the next thing you want to do is head back to

8:25your chatbot and type next, since we've now covered both the script and the voice over. At this point, your chatbot is going to ask you if you've got a visual flow for the animation in mind. Basically, do you have a rough idea of how your scenes should play out? And a visual flow is just your sequence of actions, like how the video should visually unfold. For example, you could start with a guy in bed, then he wakes up, puts on his shoes, goes outside, drinks milk, gets into his car, drives

8:51to the gym, and then enters the gym. That's all it is. Just a rough story of what happens visually. And when you give Gemini something like that, it's going to turn it into detailed prompts you can use to bring those scenes to life. Now, mapping out a visual flow is actually super simple. If you've got a rough sketch, you can just send it straight to your chatbot. But, me personally, I like to go more detailed. So, I usually build out a full breakdown with timestamps and everything. So, I'm going to send that into Gemini. And once I do that, Gemini starts giving me a really detailed video

9:19generation prompt that I'm going to use to create the visuals. Now, these prompts are made for multi-shot scene generation. So, basically, you give the model an image of your character, then you paste in the prompt, and it generates a video with multiple shots while keeping the character consistent the whole time. And that's exactly what the chatbot is doing right now. Now, since some video models have limits on how long a single clip can be, the chatbot also splits everything into chapters. So, instead of one long prompt, you get chapter one, chapter two, and so on. So, you just take chapter one, generate your first clip, then chapter two for the next clip, and

9:53you just keep going like that until your whole video is done. And honestly, that part is really, really powerful. So, right now we've got our full video generation prompts ready. The next problem is finding a platform that gives us unlimited video generation credits. And I've looked everywhere, but the only platform I found that actually does this properly is Google Flow. Now, I know Google Flow charges about 20 credits per video generation and gives you around 200 credits every single day. That basically works out to about 10 video generations per day. Now, in theory, you could create multiple accounts and multiply that limit, but that's really the only workaround here. I did look

10:30into other models that also offer free generations, but honestly, the outputs just weren't good. Like, really poor quality, and I wouldn't want to mislead you by recommending something like that. So, what we're going to do instead is just stick with Google Flow. Just head over to the platform and sign in with your Google account. I believe Google Flow gives you access to Google VO3, which is one of the most advanced video

10:51generation models available right now. And that's a big reason we're using it. Once you've signed up on Flow, just click on create new project, and it should take you to a page that looks something like this. Now, one thing I didn't mention earlier is that Google Flow also gives you access to an image generation model. So, inside Google Flow, we actually get Nano Banana at no cost, which means we can generate an unlimited number of images. And honestly, that's a really powerful addition because now we've got both an image model and a video model inside the same platform. So, I'm going to switch over to the video model. Then, what I'm

11:24going to do next is upload the image I created of my main character. Once that's done, I'll head back to Gemini and copy the chapter one prompt for this specific animation. Now, if you've got a few credits to spare, I would always recommend using C dances because it tends to produce much higher quality outputs compared to a lot of other models out there. But for me, I'm just going to go ahead and send it. Now, after a couple of seconds, we get an output that looks like this. And as you can see, it's an animated multi-shot video of our character doing different actions. I actually generated this

11:54multiple times using the exact same prompt because I wanted to see which version would give the best result. And like I said, since this is still a free model and a relatively lightweight one, sometimes you'll get small mistakes in the output. But this one actually looks really good. I like how the animation turned out and I like that there's no weird weight distortion. It also stays consistent with the character style we're going for, which is really important. Now, there are still a few slight distortions here and there, but

12:20overall, this looks solid. decided to start training. Great choice. But now the big question, calisthenics or gym? So, what I'm going to do next is just download this video. Once that's done, I'll go back and copy the chapter two prompt, paste it in along with my character image, and I'll just repeat this process until I've turned all my prompts into visuals. Right now, we've got about four chapters. [music] So, that means I only need four batches of generations to complete the entire visual creation process. So, I'm just going to run that and I'm going to speed

12:49this part up so I don't waste your time. All right. Now, once you're done animating everything, all you need to do is head back to your chatbot and type next. When you do that, the chatbot is going to give you thumbnail prompts. And these prompts are actually detailed enough to help you generate really high-quality thumbnails for this exact video. So, what you're going to do now is just copy the prompt, then go over to an image generation platform. Now, it looks like the first platform we used

13:12for reference images adds watermarks. So, if you still want to use it, that's fine. But I'd honestly recommend using Flow instead because if we go back into Flow right now and switch from the video model to the image model, we get a bunch of free models we can use. From there, you can set it to generate like four outputs. Pick your aspect ratio, paste in your prompt, and then upload your character image so everything stays

13:34consistent. And that's really important. It keeps your character looking the same across your whole brand. Once that's done, just send the prompt and Flow will generate your thumbnails without needing any other platform. So, yeah, Flow is really solid for this step. But, if you still want to use the other image tool I mentioned earlier, that's totally fine, too. So, while this is generating, I'm just going to go back to my chatbot and copy prompt variant number two. Because for the thumbnails, they usually give you a few different variants. That way you've got more options to choose from

14:02when you're making your final thumbnail. All right, right now, as you can see, we've already got a couple of really high-quality thumbnails. Our character stays consistent, which is exactly what we want. There's a bit of distortion here and there, but honestly, that's fine. That's pretty normal with free models. And sometimes you actually get one or two outputs that come out this clean. And yeah, it looks really good. I like how ready they are. You could literally upload this straight to your YouTube channel and use it as your thumbnail without doing any extra editing. I like the text, the font, the

14:29colors, even the shading it adds. Everything just looks really solid. So, at this point, all I've got to do is just download the thumbnail and attach it to my video. Now, once you're done with the thumbnail, all you really need to do is compile your video. And honestly, this part is pretty straightforward because the model has already done most of the hard work for us. At this point, we're just syncing up the voice over and the clips. Now, one thing you've got to keep in mind is that some of the clips already come with

14:52synced audio of the character talking. So, just be aware of that when you're adjusting everything. So, right now, I'm just going to create a new project in CapCut. Once that's done, I'll import the folder where I keep all my assets because when I'm working on a project like this, I like to keep everything clean and organized. So, in that folder, I've got the voice-overs, the images, and all the video clips for that project. I'll bring that into my workspace now. Then, I'm going to drag the voice-over down into the timeline and just listen through it to make sure everything sounds clean with no distortions. After that, I'll start

15:21dropping the clips onto the timeline in the exact order I generated them. Now, when generating your footage, if you end up with a few distortions, that's completely fine. All you've really got to do is crop out the distorted parts and make sure you keep the important sections intact. Now, at this point, I'm done syncing everything, so I'm just going to go ahead and export mine as a video file. And with that done, we can now move on to the next stage, creating and setting up our channel branding. All right, so for the channel creation process, we're going to head back to Gemini. If you scroll all the way to the

15:51top, you'll see where Gemini already gave us a few channel name recommendations and branding options. We've got names like Green Titan, The Green Coach, Green Physique, Prime Green, and honestly, I like all of these ideas. So, I'm just going to pick one of them. I like the names, so I'll copy one, head over to YouTube, click on my profile icon, then click create new

16:12channel, and paste in the name I want. After that, I'll just click create, and just like that, we've created our channel. Now, the next step is branding it properly, adding a logo, a banner, and writing a channel description. Now, if we go back into Google Gemini and scroll down to the channel description section, we'll see it already gave us a recommended description. We actually have two variants there, so you can just pick whichever one you prefer. But for now, let's focus on the logo. So, what Gemini is saying is we should upload our character as a reference before pasting the prompt into the image model, and

16:43that's really important for consistency. So, I'm going to copy the prompt, head back to Google Flow, and create a new project. Then I'll upload my character as the reference image. After that, I'll set the aspect ratio to 1:1, like a square, and paste in my prompt to make sure everything comes out exactly how we want it. All right, now as you can see, we've got a couple of image outputs, and they actually look really good. So, the next thing I'm going to do is take this into Canva, so we can properly crop and adjust it for YouTube. I'll log into Canva, click the create button in the

17:12top left, and search for YouTube logo. I'll click that, and it opens up a blank canvas for me. This is exactly what we want. So, I'll upload the images from Google Flow, drop them into the canvas, and then set one as the background so it fills everything properly and looks clean. Once that's done, I'll just hit share and download the final logo. Now, once the logo is ready, I'm just going to go back to the search bar, type YouTube banner, and click on one of the options. That'll take us into a blank canvas like this, and this is where we design the banner. Now, one thing about

17:42YouTube is it doesn't show your banner the same way on every device. What you see on mobile is different from what you see on a laptop, and even different from TV. So, each device basically crops it differently. To help us handle that, I already made a template that shows which areas are actually important. So, I'll just click on it, add it to the canvas, then right click and set it as the background image. Now, you'll see the red section, that's the important zone, the safe area. Anything inside there is what everyone will see no matter the device. The outer parts are mostly for TV or bigger screens. So, we just want

18:17everything important to stay inside that middle area. For me, I'm going to keep it simple and just add something like subscribe or gym tips. I'll bold it, keep it minimal, because I don't really like cluttered banners. So, I'll just type gym tips, change the font to something cleaner, and keep it nice and simple. And then from there, we just make sure the colors match between the logo and the banner so everything feels consistent. If you want to build a brand that actually looks professional and worth investing in, you've got to make sure your color theme stays consistent across everything, your banner, your logo, and even your thumbnails. So,

18:51since we've already created our logo, I'm going to bring it into this canvas just so I can pull colors directly from it. First thing I'll do is select the text on the banner and match its color to the exact green from the logo. So, I'll click on the text, open the color wheel, and I'll see a bunch of color options. But, instead of guessing, I want the exact same shade that's in the logo. So, I click on the logo image itself, then open the color picker there, and it gives me the precise color code for that green. I'll copy that code using control plus C. Then I go back to

19:20my text, open the color wheel again, paste in [music] that exact code, and click change all. And now you can see the text color matches the avatar perfectly. It's the exact same green. I'm also going to do the same thing for the background. So, I'll just click on my logo image, copy the exact background color, go back to the banner, and paste that same color in so everything blends properly. Once that's done, I'll remove the logo from the canvas, and boom, the banner is done. So, I'll just share it and download it. Now I'm going to head back to YouTube, click on customize channel, which takes me straight into

19:51YouTube studio. Then I'll scroll down, upload the logo I just created, and also upload the banner I just made. Next is the description. So, I'll go back into Gemini and just copy one of the description variants. I really like how short and punchy it is. And this is actually something a lot of people get wrong. They write these long story-style descriptions. But honestly, keeping it short and clean is way better. Someone should be able to instantly understand what your channel is about. So, I'll paste that in, maybe add one emoji just to keep it simple, and then click publish. Now when I go back to the

20:23channel, you can see everything looks really clean. I like how consistent the colors are, the palette looks nice, and the minimal style just works. And I also like how the avatar really feels like the face of the brand now. Everything just looks really solid.