How to make an AI video
Six steps, with a picture of every screen. It takes about five minutes the first time and under a minute once you know where things are. You can read the whole thing without an account.
Open the studio
Go to Create. Everything happens on one screen - the box you type in, the models you can choose, the length and shape of the clip, and a wall of ready-made looks. Nothing is hidden behind a wizard and you can look at all of it before signing in.
You do not need an account to look around. You need one to press the button, and a new account starts with 35 free credits and no card.
Say what you want, or upload a photo
Type what the video should show in the box at the top, or use the upload button on the left to start from a photo. A photo is the stronger start: the models keep the subject recognisable and animate around it, which is how most people use the site.
Write it like a shot, not a wish. “Slow push in on a woman turning to face the camera, rain on the window behind her, evening light” gives you far more than “a cool cinematic video”. Say what the camera does first - it is the instruction the models weight most heavily.
Choose a model
Pick the model that matches the job. They differ in price, in length, and in what they are good at: some follow camera instructions precisely, some generate spoken dialogue and sound with the picture, and the cheap ones are for trying an idea before spending on it.
A speaker icon on a card means that model makes sound. The price changes with the model, and the button always shows what this render will cost before you press it.
Set the length, the shape and the quality
Use the column on the left, or the Settings button if you are on a phone. Duration is in seconds, dimensions decide whether the clip is wide for a screen or tall for a phone, and quality runs from 480p to 1080p. Longer and sharper costs more, and the cost updates as you change them.
For anything going on Instagram, TikTok or Shorts, choose Portrait 9:16. For YouTube or a website, Landscape 16:9. Getting this right at the start saves you cropping a shot somebody framed properly.
Or start from a template
If you do not want to write anything, click a template. Each tile is a finished effect - the prompt, the model and the framing are already set, and the clip on the tile is what that template actually produces. Add your photo and press the button.
What is on the tile is what you get. A tile showing a photograph sets up a photograph, a tile showing a clip sets up a clip.
Press Generate, then find it in your Library
The render takes somewhere between thirty seconds and a few minutes depending on the model and the length. You do not have to wait on the page - every finished clip lands in your Library, where you can play it, download it, or send it back into the studio to try again.
Credits come off when the render succeeds. If it fails, you are not charged. You can see what everything costs on the pricing page.
Questions people ask
What is buildmy.video?
buildmy.video is an AI video generator. You type what you want or upload a photo, pick a model, and it renders a video - cinematic clips, product shots, social ads, or anime scenes with your own characters. It runs 6 different AI models behind one credit balance, so you are not paying a separate subscription for each one.
Can I turn a photo into a video?
Yes. Upload a photo and the models animate it - camera moves, motion, lighting changes and effects, with the subject of the photo kept recognisable. This is the most common way people use the site: one picture in, a short clip out.
How much does it cost?
You start with 35 free credits, no card. After that you buy credits or take a plan: Starter is $19 a month and Pro is $49 a month. Every model shows its credit cost before you press the button, so you always know the price of a render before you spend anything.
Is there a free AI video generator here?
Yes, to a point. New accounts get 35 credits with no card and no trial countdown, which is enough to make real clips and see whether the output is any good. There is no free unlimited tier - the models cost real money to run and we would rather say so than hide it.
Which AI video models can I use?
6 models, including Seedance, Kling and Veo, plus image models for stills. You pick per render rather than being locked to one, because different models are genuinely better at different things - camera moves, speech, realism, or price.
How long can the videos be?
Most clips run 5 to 10 seconds, and the flagship model reaches 30 seconds in a single render. Short clips are what the current generation of models does well; anything longer is several shots joined together rather than one continuous render.
Can I use the videos commercially?
Yes. What you generate is yours to use, including commercially. You are responsible for what you put in - do not upload footage, music or faces you have no right to use, and the site will ask you to confirm the rights before it renders anything that looks like somebody else's work.
Do the videos have sound?
Some models generate sound and spoken dialogue with the picture, at no extra charge. Others are silent and you add audio afterwards. The model list says which is which before you choose.