How to make a picture move with AI: pick the motion

To make a picture move with AI, give it to an image-to-video model as the clip's first frame and describe the motion in your prompt; the model generates the seconds that follow. To copy a movement from another video, use motion control; to make a face speak, use lip sync.

The steps below use Sume, where one image-to-video clip runs 2 to 30 seconds depending on the model. Facts come from the Video generation and Models overview docs, the OpenAPI document behind the Sume API reference, and Sume's video catalog, read on 2026-09-27.

Which AI tool makes a picture move the way I want?

It depends on where the movement comes from:

How do I animate a photo with image-to-video?

• Put the picture at a public HTTPS URL. Localhost, private-network, non-HTTPS, and signed or private URLs are refused before the job is submitted.
• Pick a model. Every video model accepts a first frame, and grok-imagine-video-1.5 does image-to-video only.
• Describe the movement in prompt. The docs suggest details about motion, camera angles, lighting, and scene composition.
• Set duration inside the model's range: seedance-2.5 and wan-3.0 go to 30 seconds, and the rest stop at 15 or less. AI video length limits by model lists them.
• Poll the job until status is completed, then download the clip with your API key.

curl -X POST https://api.sume.com/v1/videos \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: photo-move-001" \
  -d '{
    "model": "wan-3.0",
    "prompt": "The woman in the photo turns toward the camera and smiles as wind moves her hair",
    "frame_images": [
      {
        "type": "image_url",
        "image_url": { "url": "https://example.com/photo.jpg" },
        "frame_type": "first_frame"
      }
    ],
    "resolution": "720p",
    "duration": 5
  }'

Can I copy the movement from another video?

Yes, with Kling 3.0 Motion Control. It animates a still with the motion of a driving video you supply, and the output length follows that video, which can be up to 30 seconds. The optional prompt only steers appearance details; motion and framing come from the driving video. Fields and price are in Motion control API: animate an image with a driving video.

Can I make the person in the photo talk?

Not with image-to-video: Sume's docs say video models do not lip-sync to generated speech or to a later voice-over. Use VEED Fabric 1.0 instead. It turns a still plus Sume-hosted audio, such as a text-to-speech clip, into a talking clip. The lip sync API guide covers the request.

What are the limits?

• A still on Sume's editing timeline doesn't move: in a Timeline 1.0 render, a still is a static hold, and its motion field is accepted and ignored.
• Only the first frame is your picture. The frames after it are generated, and no setting is documented that keeps a face or detail identical, so watch the clip before you use it.
• The picture must be at a public HTTPS URL. A signed link or a file on your own computer won't work.
• One clip tops out at 30 seconds. For longer footage, generate several clips and join them.