A clean subject
One clear subject with space around it gives the model room to add camera movement without artifacts.
Animate image to video
Animate Image to Video: Still Frame, Moving Clip
Animate image to video in one pass: the still you upload becomes the first frame, the prompt you write becomes the movement, and the model returns a clip you can download. The aspect ratio, duration and credit cost are set before the job runs.
This is the same image-to-video workflow that powers the studio, documented end to end so you can judge the steps and the cost before you spend credits.
Free tier: 50 credits, no credit card required
How image to video animation works
- 1
Choose the first frame
Upload a JPG, PNG or WebP still. The model starts from this image, so framing it the way you want the clip to open saves rework.
- 2
Write the movement
Describe what changes over the clip: a camera push, a turn, wind, water, a hand gesture. Keep it to one or two clear actions per run.
- 3
Pick the model and duration
Cheaper models suit drafts; Sora 2 and Wan 2.5 cost more but handle longer, more detailed motion. The credit cost is shown before you run it.
Good first frames
A still with depth
Landscapes, streets and interiors respond well to slow push-ins and parallax.
A character you need to keep
Reference-image runs keep the face or product shape recognizable across the clip.
A frame you will reuse
Keep the source still if you plan to try several models, then compare the clips side by side in your history.
What each model costs
Every run shows its credit cost before you confirm, so you can compare models without guessing.
| Model | Credits per run | Best for |
|---|---|---|
| Flux Kontext | 30 | Cleaning up or restyling the still before you animate it |
| Veo 3 | 300 | The cheapest short clips from a prompt or a reference image |
| Kling | 500 | Character and object motion from a single frame |
| Seedance | 600 | Fast image-to-video drafts |
| Wan 2.5 Image to Video | 800 | Animating a reference still while keeping the subject recognizable |
| Sora 2 | 1,000 | Longer cinematic clips with strong camera motion |
Image to video FAQ
- What is the difference between image to video and text to video?
- Text to video starts from a written scene and lets the model design the visuals. Image to video starts from a picture you provide, so the composition and subject stay close to the original.
- How long are the clips?
- The studio offers the durations each model supports, and the choice is made before the run — longer clips and heavier models cost more credits.
- How much does one image to video run cost?
- Veo 3 starts at 300 credits, Kling at 500, Seedance at 600, Wan 2.5 Image to Video at 800 and Sora 2 at 1,000. Every run displays its cost before you confirm it.
- Which model should I start with?
- Start with a cheaper model to test the prompt, then move to Sora 2 or Wan 2.5 once the motion looks right.
- Where do the results go?
- Finished clips stay in your generation history so you can download them again, and any failed run is refunded in credits.
Related tools
Animate a photo, try animate image AI, or check the pricing page before you scale up.
Turn an image into video
Upload the frame, describe the movement and run the job. Costs are shown up front.
Open the studio