What CASTMINT does

CASTMINT turns a written prompt or script into a short vertical video, ready to post to TikTok, YouTube Shorts or Instagram Reels. You write the idea; the finished MP4 comes back a few minutes later.

How it works

Every video goes through the same five steps:

1. Script. Your prompt becomes a short narration script. If you already have a script, you can paste it instead and skip this step.
2. Scenes. The script is split into scenes, each with its own line of narration.
3. Images.An image is generated for each scene from a visual description derived from that scene's text.
4. Voiceover. The narration is read by a synthetic voice you pick from a fixed set.
5. Composite. The scenes, the voiceover and the captions are rendered into a single MP4. Caption timing is cut to the measured timing of the recorded speech, so the words on screen match the words being spoken.

What you get

An MP4 you download and own, at the aspect ratio you chose. Videos on paid plans carry no CASTMINT watermark.

You also get an in-browser editor for the videos you have generated. It covers captions (on/off, style, wording per scene, and position on the frame), text and image overlays, the order of the scenes, how long each scene holds relative to the others, and the choice of voice. Changes are re-rendered into a fresh MP4 when you export.

Who it is for

Creators and small businesses that need a steady supply of short-form video and do not want to run a camera, a microphone and an editing suite to get it. No footage, no studio and no editing experience are required.

How it is sold

CASTMINT is sold as a monthly subscription. Each plan includes a monthly allowance of credits, and generating a video spends credits — a longer video costs more than a shorter one, and some options (a premium voice, a generated music bed) cost a little extra. Credits are spent at generation time, so you always know what a video costs before you make it. Current plans and prices are on the pricing page.

The AI behind it

CASTMINT is an interface built on top of third-party AI models. We do not train our own models. Separate providers handle script writing, image generation, speech and music, and which providers we use may change as the field moves — what stays constant is the pipeline above and what you get out of it.

What CASTMINT does not do

Some things a video generator could plausibly offer, we deliberately do not build. These are limits in the product, not just in the terms:

• No face swapping.There is no way to put one person's face into another person's footage. The feature does not exist in the product.
• No voice cloning.Voiceovers use a fixed set of licensed synthetic voices. There is no path to upload a recording and reproduce a particular person's voice.
• No likeness without consent. A photograph of a person can only be used after you confirm that it is your own image, or that the person in it has given you explicit permission to create AI video from it. That confirmation is required by the service before the photo is accepted, and it is recorded.
• No adult content. Sexual and sexualised content is out of scope and prohibited.

Content and conduct

You are responsible for what you generate and for where you publish it. Prompts are screened automatically before generation, and we may act on content that breaches our Terms and Conditions— including impersonation, deceptive use of a real person's likeness, and material that infringes someone else's rights. Where you connect a social account, publishing goes to that account and you remain bound by that platform's own rules.

See also our Terms and Conditions, Refund & Cancellation Policy and Privacy Notice.