Generating video
Make a short clip from a prompt or a start frame with Seedance, Veo or Grok, and place it as a video layer or a scene.
AI → Generate video… makes a short clip from a description, and can animate a picture you already have. It uses your Hypery sign-in and your AI credits, the same way Generate image does.
Making a clip
Description
Signed in, in the editor. AI → Generate video… opens the dialog. A description is typed — a paper boat on a rainy street, the camera following it — and the model, length, shape and resolution are shown, ready for Generate.
- Open AI → Generate video…, or type "Generate video" in the command palette.
- Describe the clip: who or what is in it, what happens, and how the camera moves. For example, "A paper boat drifting across a rainy street at dusk, slow dolly shot".
- Choose a model, a duration, an aspect ratio and a resolution. Each model offers only the options it supports.
- Click Generate.
A placeholder appears in the middle of the view with your prompt underneath it, and you can keep working. A clip takes from about 30 seconds to a few minutes. When it's ready it plays in the placeholder, which takes the clip's exact shape.
The dialog lists the videos you've started this session, with their progress and how long they've been running. Stop cancels one. So does deleting its placeholder. A model that has already started a video may still charge for it.
If a generation fails, the placeholder turns into a red card that says why, such as a prompt the model refused or running out of credits.
Starting from a picture
Select one or more layers and tick Start from the selection. PopCraft renders them as a single picture, the same way it exports them, and the model animates from that picture. Describe the motion rather than the scene.
To continue a video, select the video layer. The frame it's showing becomes the first frame of the new clip.
Seedance models, Veo 3.1 Fast and Grok Imagine can start from a picture. Veo 3.1 Lite works from text only.
In Animate
When a composition is open in Animate, a finished clip becomes a new scene at the end of it. The new scene is a frame the size of the composition's stage, with the video filling it, and its timeline lasts as long as the clip. You can trim, reorder and add transitions to it like any other scene.
Models
| Model | Durations | Resolutions | Sound | Start frame |
|---|---|---|---|---|
| Seedance 2.0 Fast (default) | 4–15 s | 480p, 720p | Yes | Yes |
| Seedance 2.0 | 4–15 s | up to 1080p | Yes | Yes |
| Seedance 2.0 Mini | 4–15 s | 480p, 720p | Yes | Yes |
| Seedance 2.5 | 4–30 s | up to 1080p | Yes | Yes |
| Seedance 1.5 Pro | 4–12 s | up to 1080p | Yes | Yes |
| Seedance 1.0 Pro Fast | 2–12 s | up to 1080p | No | Yes |
| Veo 3.1 Fast | 4, 6 or 8 s | 720p, 1080p | Yes | Yes |
| Veo 3.1 Lite | 4, 6 or 8 s | 720p, 1080p | Yes | No |
| Grok Imagine Video 1.5 | 1–15 s | up to 1080p | Yes | Yes |
The dialog shows an estimated cost in credits, based on the model's list price. The final charge comes from your Hypery account. As a guide, a 5-second 720p clip is about 10 credits with Seedance 1.0 Pro Fast and about 60 with Seedance 2.0 Fast. An 8-second Veo 3.1 Fast clip with sound is about 120.
Where the file lives
A generated video is stored with the file's local media, just like a video you drag in. It's saved on this device, works offline, and syncs for cloud files. It is never embedded in the document itself.
For the agent
The agent can call generate_video (the generate.video command) with the same options as the dialog:
prompt, model, duration, aspect, resolution, audio, startFrameIds (the layers to render as the
start frame) and composition (add the clip as a scene). Without composition, the clip is placed as a video
layer at x/y, width wide.