Start with a strong image
A clear subject and uncluttered composition give the model a useful starting frame. Avoid tiny faces and cropped limbs.
Image to Video AI
Bring a photo into motion. Describe the action and camera, choose your settings, and create a video with native model audio.
Create your video
Opening your workspace…
How to turn an image into a video
Upload one clear photo
Use a JPG or PNG up to 10 MB. Keep the subject in focus and leave room for movement.
Describe the motion
Tell the model what should move and how the camera should follow. Choose a model and output settings.
Review, generate and download
Check the credit quote, confirm generation and follow the task here. Download the finished MP4 from your private works.
Give your photo a clear direction
Describe movement, not the whole picture
Focus on one subject action, camera movement and atmosphere. Keep important features you want preserved explicit.
Choose a matching frame
Portrait, square and landscape are available. A ratio close to your original photo can reduce unwanted reframing.
Credits follow your settings
Your model, resolution and duration set the quote. Review it before confirming a generation.
Image to video AI: your questions
What image formats can I upload?
Upload one JPG or PNG up to 10 MB. Use a photo you own or have permission to use, and obtain consent from anyone shown.
Will the subject stay exactly the same?
The photo supplies a starting frame. Identity, anatomy and small details can change; inspect the result before publishing.
Which models and settings can I choose?
Seedance 2.0 Mini and Fast support 480p/720p; Standard and 2.5 also support 1080p. Choose a whole-second duration from 5 to 15 seconds.
Will the video include sound?
New formal-site tasks request native model audio. It is generated for the scene, not copied from another video or guaranteed to reproduce a song.
How are credits charged?
Credits vary with the selected model, resolution and duration. Review the quote before confirming. Confirmed technical failures and uncertain results are handled separately.
Where can I find and download the result?
The task appears here and in My creations. Private media is retained for 30 days. If a task is pending or under review, follow that task rather than submitting it again.