Animate a still, or start from a prompt
Two routes into the same output. Image-to-video takes a product still and adds motion - a slow push in, a pan, a rotation - keeping the product itself intact. Text-to-video builds the shot from a written description when you have no footage to work from. Most video ads for physical products come out better from the first route.
Sized for the placement, not cropped afterwards
Vertical for Reels and TikTok, square for feed, landscape for pre-roll. Setting the aspect ratio at generation time means the composition is built for that frame, rather than a wide video cropped down until the product falls out of shot.
Length that matches the plan, cost that matches the run
Video generation is the most compute-heavy thing on the platform and it is priced accordingly. Cost is per 5 seconds of output, from 45 credits a block, so length drives price directly - a 10-second cut is two blocks, a 30-second one is six. The exact figure is shown before you confirm. Pro generates up to 30 seconds and Studio up to 60.
Part of a chain, not a dead end
A generated video is not the last step. Pull a frame back into image editing, upscale it, or convert the product into a 3D asset for a rotating shot. The whole chain runs on one credit balance in one session.
Frequently asked questions
- How long can the videos be?
- Up to 30 seconds on Pro and up to 60 seconds on Studio. Basic plans generate shorter clips. Most social ad placements sit well under those limits.
- Can I add my own text or logo over the video?
- Overlay text can be described in the prompt, but for exact brand lockups and legal copy you are better off exporting the clip and adding those in your own editor, where you control font and placement precisely.
- Is there a watermark?
- Not on paid plans. Free trial credits produce watermarked output so you can judge quality before subscribing.
- Which is better for a product ad - image-to-video or text-to-video?
- Image-to-video, in almost every case. Starting from a real photo of the actual product keeps it recognisable. Text-to-video is the better choice for abstract or atmospheric shots where no specific product needs to stay accurate.
Related tools and workflows
Ready to create with AI?
Start with free trial credits. Pay only for the generations you run.
Generate a Video Free →