Transparency · 14 August 2026 · 4 min read
What an AI-generated video actually costs to make
Real measured numbers for a ten and twenty minute video, stills and animated, with nothing hidden.
Most AI video tools will not tell you what a video costs them to produce. We would rather explain it, because the numbers are the reason the pricing is shaped the way it is.
Where the money goes
Four places: the language model writing the script, text-to-speech for narration, image generation for every scene, and the final render.
Image generation dominates. A twenty-minute video is roughly ninety scenes, and every one of them is an image.
Why animation triples it
An animated clip is not one image, it is a short video generated from one. That is a fundamentally heavier operation, and it runs on more expensive hardware for longer.
Hence the pricing: an animated minute costs three credits and a stills minute costs one. That ratio is not a margin decision, it is roughly the actual cost difference.
Why there is a spending cap
Cost scales with length and scene count, so a thirty-minute animated video is several times a five-minute one. A per-project cap means that is a decision you make deliberately rather than discover afterwards.
See it applied to a real video.
Every example in the showcase includes the sentence that started it and every stage in between.
Open the showcase