Bring a portrait to life
“Animate this portrait: she blinks, smiles softly and turns her head a few degrees toward the camera. Keep the lighting and background unchanged, subtle natural motion only.”
Upload a photo as the opening frame, describe the motion, and EveryGen AI turns it into a video with sound. Add a closing frame to decide where the shot ends.
EveryGen AI image to video animates a still image into a short clip. The image becomes the first frame, your prompt says what moves, and the video agent writes the detailed shot prompt, picks the model and renders the clip with sound.
It works with photos, illustrations, product shots and AI-generated stills. Clips run 4 to 30 seconds depending on the model, and every take stays in the same chat, so you can ask for a slower camera or a different ending without starting over.
| Input | An opening frame, plus an optional closing frame. JPG, PNG or WEBP, up to 100 MB each. |
|---|---|
| Models | MiniMax H3 and Seedance 2.5 |
| Clip length | MiniMax H3: 5–15 seconds; Seedance 2.5: 4–30 seconds |
| Resolution | MiniMax H3: 768P / 2K; Seedance 2.5: 480p / 720p |
| Aspect ratios | 21:9, 16:9, 4:3, 1:1, 3:4, 9:16, or matched to your image |
| Sound | Generated together with the picture on every model |
| Cost | MiniMax H3 at 2K: 100 credits per 5 seconds; Seedance 2.5 at 720p: 400 credits per 5 seconds |
| Output | MP4 video with audio, saved to the chat and your library |
Drop a JPG, PNG or WEBP (up to 100 MB) into the composer. Add a second image if you want the clip to end on it.
Say what should move and what the camera does, for example “slow push-in, hair moving in the wind”. A short line is enough; the agent fills in the detail.
The clip renders with sound in a few minutes. Ask for changes in the same chat and every version is kept.
Upload your image first, then use one of these as the prompt.
“Animate this portrait: she blinks, smiles softly and turns her head a few degrees toward the camera. Keep the lighting and background unchanged, subtle natural motion only.”
“Animate this photo with a slow cinematic push-in on the subject, gentle hair movement in the breeze, background softly falling out of focus.”
“Animate this still with a 2.5D parallax move: the camera drifts slowly to the right while clouds move and light shifts across the scene.”
“Animate this product photo: the camera orbits slowly around it while a soft light sweep crosses the surface. Keep the background clean and the product shape unchanged.”
“Animate this pet photo: it tilts its head, ears twitch and it looks into the camera. Natural fur movement, soft daylight, slight handheld feel.”
“Use the first image as the opening frame and the second as the closing frame. Move smoothly between them with a slow dolly forward and a gentle change of light.”
Yes. Upload the photo as the opening frame, describe the motion, and EveryGen AI renders a clip from it with sound. You can add a second image as the closing frame.
JPG, PNG and WEBP, up to 100 MB per file. Up to two images are used as frames: the first as the opening frame, the second as the closing frame.
It depends on the model: MiniMax H3: 5–15 seconds; Seedance 2.5: 4–30 seconds. You choose the length in the composer before you send.
Yes. The image is the first frame, so the subject, colours and composition carry into the clip. Describe only what should move or change.
Only photos you have the right to use. Upload images you own or have permission for, and do not animate someone's likeness without their consent.
The price depends on the model, length and resolution, and the composer shows it before you send. A 5-second MiniMax H3 clip at 2K costs 100 credits. Failed renders are refunded automatically.
Upload a photo, describe the motion and get a clip with sound.