In this video I show you how to make your first AI video on Higgsfield AI as a complete beginner. No camera, no experience. I make a character with GPT Image 2.5, build a character sheet so the character stays the same in every shot, and turn it into a realistic AI video with Seedance 2.5. One AI video generator, every model in one place.
⏱ Chapters 0:00 What can a beginner make with AI video today? 0:12 Intro 1:12 Which AI model should a beginner use on Higgsfield? 2:40 How do you make a character with GPT Image 2.5? 3:50 How does a character sheet keep the character the same? 4:46 How do you make a video with Seedance 2.5? 5:25 How to write the perfect AI prompt? 7:00 My first AI video attempt. 8:56 How can I remix videos with AI?
📝 My prompts Character: Cinematic portrait of a weathered Viking warrior in his 40's. He has a short beard. Leather and steel plated armor. his face is dirty and has a scare across his forehead. Frame is chest up. 85mm lens with f/2.
Character sheet: Character sheet, three panels of the same person from the attached reference: full-body front view, chest-up three-quarter portrait. Identical face, skin tone, hair, body proportions and outfit in all views: back view shows the same hairstyle and back of the garments. Neutral standing pose, relaxed empty hands, full figure head to shoes, seamless white studio background, soft diffused lighting, Unretouched natural skin, fine lines, no make up, matte finish. Photorealistic, high-end commercial photography.
CHARACTER REFERENCE: Use @image of Viking character sheet as the exact visual reference for the protagonist. Maintain his facial features, short dark hair, greying beard, facial scar, weathered appearance, dark layered leather and iron armour, fur-covered shoulders, brown wool cloak, belts and rugged Viking clothing throughout the entire sequence. Preserve his identity, proportions and costume across every shot.
SETTING: Northern Scandinavia, deep winter. A remote, ancient Nordic forest surrounded by snow-covered ground, bare trees, frozen vegetation and towering rocky landscapes. Overcast skies, cold blue-grey colour palette, atmospheric mist and a sense of complete isolation.
Open with a cinematic wide shot of a remote northern Scandinavian forest in the middle of winter. Tall, bare trees sway slowly in a cold, powerful wind. Their skeletal branches creak and move against a pale, overcast sky. Fine snow particles drift through the air.
The camera slowly glides between the trees. The atmosphere is hauntingly beautiful, silent and desolate.
AUDIO: Begin with eerie ambient forest sounds. Cold wind passing through branches, subtle rustling foliage, distant creaking trees and a deep atmospheric drone. No music initially.
Cut to an intimate extreme close-up of the Viking lying motionless on the cold forest floor, partially surrounded by snow and dead leaves.
Suddenly, his eyes snap open.
He inhales sharply, as if violently awakened from a vivid dream. His expression is confused, alert and troubled. His breathing is visible in the freezing air.
The camera slowly pulls back as he raises his head and begins getting up.
AUDIO: A sudden sharp inhale, followed by the subtle sound of leather and armour shifting.
Medium tracking shot as the Viking gets to his feet. He looks around cautiously, scanning the surrounding forest.
Cut to his point of view: an endless landscape of ancient trees, snow-covered rocks and distant mountain silhouettes disappearing into the winter mist.
There is nobody. No movement. No sign of life.
He stands completely alone in the wilderness.
Use slow, deliberate camera movements to emphasize the vastness and isolation of the environment.
AUDIO: Wind, distant branches creaking, his heavy breathing. The ambient drone gradually builds.
The Viking suddenly freezes.
From somewhere deep in the distance, a woman screams.
He immediately turns his head towards the sound. His expression changes from confusion to alarm and determination.
Hold a brief close-up of his eyes as he listens.
The scream echoes faintly through the forest, swallowed by the wind.
AUDIO: A distant, desperate female scream. Brief atmospheric silence immediately afterwards, followed by a rising cinematic tension.
This video provides a comprehensive tutorial on using Higgsfield to create AI-generated videos. The author demonstrates the end-to-end process, from generating consistent character references and character sheets to producing high-quality cinematic clips, while emphasizing the importance of AI skills for modern content creators and filmmakers.
So I created this video in just a few minutes. I'm going to show you
how you can create your first AI video if you've never tried it.
Now this might be especially useful if you're,
for example, running a production company.
You could, for example, have
an online influencer account that you're running.
Or like in my case, I like to make films.
So let's say you want to create a little test film
or sort of something to test out your concept that you have.
Or maybe you want to do a pitch for your script.
So first, you're going to start by clicking
the link in the description of this video.
We'll take you to this page.
And this kind of shows you already kind of more
or less what I'm going to show you in the video.
Now, obviously, the first thing you're
going to have to do is to create an account.
Hicksfield does every so one run like promotions.
So you can see one right now running up here.
So we'll check it out.
And they'll show you basically a few different plans.
You have obviously the free plan
that you can have with them, the starter plan.
But I would recommend that you always
go with the plus at the beginning.
Simply because that's the the cheapest plan
that you can get that will give you access to Seedance 2.5
And that's what I've used for generating my video
and that's really I would say right now kind of on the market the best
Video generation model as you can see right now
They're running a promotion $35 a month
if you do it the yearly subscription
And with that you're getting 1200 credits right now
and not only that but they're for example
giving you right now unlimited access to Seedance for 14 days
and also they're giving another 60
days of unlimited access to Nano Banana Pro.
So anyways go through this sign up for your account
with your email or whatever your preference
is and then let's jump right into the tutorial.
Once you're signed in up here you'll see
on Higgs field page you have your different tabs up here
and basically the way it works is you first want
to create your images. Images are going to be
for your let's say character references,
your locations, things like that. Now the video part
of this is basically for bringing those images to life
so you can add motion to them or put those
let's say characters in those images
in different environments, different settings.
So we're going to go to image up here
and I'm going to choose GPT image flare and the reason
is because that model just follows the prompts very precisely.
And you can basically start typing up here what kind
of a character you want to create.
You can also load in image references.
For example, if you're working, like in my case,
let's say a lot of times I'm working
with an actor in a film and I want to, let's say,
create an extra scene or shots
or things basically that we weren't able to film in real life,
but with my actor actually being in here.
For example, let's say you want to create a little
Viking story or action scene.
So let's say we're going to create our Viking character here.
Now by the way if you need help with creating
your prompts you can just copy and paste my
prompts. I'm going to put them in the description of the video
but you can also use for example
ChatGPT or Claude to help you actually write kind
of more detailed prompts.
And there's our image of our viking. Looks pretty cool.
Now for my video, like I said,
I would normally want to sort of put my actor in it.
In this case, it's a test. So I'm just going to put myself in it.
And I'm going to do that by loading references of my face.
And I can do that up here.
Just click plus and choose the images that you have.
And we're going to click generate.
I mean, it looks kind of cool, but I think my head, I don't know,
seems like it's not very proportional.
So maybe I'm going to go in there.
I'm going to kind of tweak my my prompt a little bit.
And I'm just going to create a few more kind of,
just so I have a few different options to pick from.
And there, I actually like this image quite a bit.
So I'm going to use this basically as a reference.
You can click up here, loads it into here.
I can, you know, delete the other references.
And I'm going to clear my prompt here
and now what we're going to do is create a character sheet
which essentially shows you your character from different angles,
full frame, kind of full body
and obviously the close-up
and this gives the AI video model a really good reference so that
it keeps the character consistent from shot to shot.
Now for this one we're going to change this
and we're going to actually choose the GPT image sound burst model
and we're also going to change
I'm going to change the aspect ratio to 16 by 9.
And then you go and click generate.
Now I'm going to create another one just to kind of see
what variations it will do.
Which by the way again even with this model you can for example
create a batch of these.
So let me maybe create for example two of these. So once you have your
character sheets you can eventually check for all
the details you want. Make any changes if
you need to and then let's say just pick you know
your favorite one. Next step is to go and
We'll go and create video.
So we'll go here to video, create video.
I'm going to click the model up here, and I want to use Seedence 2.5.
Now, there are other models.
One other one you might want to use is Seedence 2.5 Draft,
which essentially creates your video by the lower resolution,
saves you credits, so you can do, let's say, a few different versions.
And I kind of touched more on that in my last video,
where I kind of show you when I was doing a lot of different videos
and kind of testing a lot of ideas.
That's where this can save you a lot of money.
We're going to click, for example, our references up here.
We'll go to Image Generations,
and we're going to pick one of these character references.
Now the next step is to write your prompt
so the prompt is basically you're telling the video model
who's in the video what happens how the camera behaves
and also what we're going to hear.
Now there is no really I would say
a perfect prompt you can write there's a lot of different
ways you can put it essentially the more you kind
of leave out if you just do a very quick simple
prompt then the AI model is just going to in kind
of imagine a lot of other aspects of it and you
And you might just get really weird results.
Or results that you weren't really imagining.
So the more detail you put into this, the better it's going to be.
Again, you can use ChatGPT, for example,
or Claude to help you write a long, detailed prompt.
In case, for example, you're not familiar
with how to describe certain actions.
Or if you don't know camera or filmmaking terminology.
If you're actually trying out while you're watching this tutorial,
then feel free to just copy my prompt, again, in the description.
And see kind of you know what you get and then from that point
on then you can kind of go in there and tweak it.
Now one thing you'll notice in my prompt
is that I referenced the image up here so you're going to
basically just go up here you can click
on this and you're going to select your reference image
that you want to use. You're essentially telling
the model to use this again the character sheet
and otherwise obviously you can go through this and you can adjust the
prompt to however it is that you like to have it.
You can set the duration obviously to whatever you want to.
I'm going to set this one all the way to 30 seconds.
I'm going to set the quality to 720p just
to kind of save on credits as you can see.
And that's it. We click generate and here's our video.
Let's preview it.
Obviously, if you wanted to change it, for example,
let's say you wanted to add a weapon to him,
you can add all of that in there.
You can also actually add more references. So for example,
let's say you want him to have
a specific item like a weapon in his hand,
then you can add a reference image of that,
which again, you can generate up here.
And then this way that that object that in his hand
is going to stay consistent shot to shot.
Same thing, for example, for the environment,
like let's say there's a specific location
you, for example, like,
like in the real world place that you've been to take pictures, you
know, real references, or again,
you can generate them up here, and this way you can tell the
video model to always keep that location consistent.
Let's say, for example, I wanted to tweak this,
you know, for example, the lighting,
let's say I'm not happy with it,
I want it to be kind of more, let's say like a late
kind of a, kind of a twilight kind of almost sunset setting.
And you can see with the right tweak
of your prompts you can change for example the time of day,
the lighting or the you know completely for example the overall feel
and action of your scene.
Once you have
a version
that you like you
just click up here
to download it.
Now you can also have fun for example with Genjutsu.
We can essentially remix one of the trending videos up here.
And let's say for example we're gonna remix this video here.
So click recreate and we're gonna apply our own reference up here.
I'm going to change the resolution to 720p,
you can also you see use the three generations,
still have three left and go click generate.
All right let's see what we created up here.
This is pretty fun
so let me for example watch out another video here.
And for example I'll take this one,
make it look like I'm a soccer player.
Again I'm just going to add my reference image and click generate.
So as you can see, I mean, you can have
a lot of fun with this AI stuff,
but I think also like if you're,
whether you're in this industry of like, you know, filmmaking,
videography, or creating online content,
or pretty much in any other industry, I'm sure you've
noticed AI is making a lot of changes
and kind of speeding up the process, replacing a lot of jobs
too, but also creating new jobs.
And I'll tell you that if you want to kind of stay ahead,
I know for myself, a lot of the work right now that I'm doing,
And a lot of the clients that I've gotten
is because I've been developing my AI skills.
Right now, pretty much half the articles online are written by AI.
And when it comes to just video generation,
like for example, since last year, more than
a million YouTube channels are already creating AI content every day.
And so definitely, if you want to be able
to compete with online content, you kind of
have to get into AI and learn to use it.
And because these tools are now so readily available and affordable,
I think it's just going to really come down to how well
and creatively you can actually use them
and what kind of work at the end
of the day you're able to create with it.
So if you want to get started, click the link in the description.
As a new user, you're going to get 50% off.
Like I mentioned before, start with the Plus subscription
and use your first week of unlimited Nano Banana Pro
to really experiment and start learning these tools.
Anyways, that's it for this video. I'll see you in the next one. Bye!