Video model · Image to video
Seedance 2.0 I2V on Klingobanana
Animates a still image into a cinematic clip with audio. You supply the first frame, so the subject, outfit and framing are decided before any video credits are spent.
From 104 credits per clip
Specs
- Type
- Video · Image to video
- Input
- Requires an image
- Reference images
- Up to 1
- Clip lengths
- 5 s, 8 s, 10 s, 15 s
- Resolution
- 720p or 1080p
- Audio
- Native audio
- Speed
- Slower, higher fidelity
Price per generation
| Option | Cost |
|---|---|
| 5 s at 720p | 46 credits |
| 5 s at 1080p | 104 credits |
| 8 s at 720p | 74 credits |
| 8 s at 1080p | 166 credits |
| 10 s at 720p | 92 credits |
| 10 s at 1080p | 207 credits |
| 15 s at 720p | 138 credits |
| 15 s at 1080p | 310 credits |
Credits come from your plan. Planning a batch? Use the video cost calculator to see the euro cost per plan.
What Seedance 2.0 I2V is good at
Seedance 2.0 Image to Video takes a picture you already like and gives it motion and sound. Because the first frame is yours, the character, outfit and composition are settled before any video credits are spent. That is its biggest advantage over text-to-video, where the model invents everything and a good face can be lost on the next try.
This is the standard-tier version, so it shares the resolution choice and the higher cost of Seedance 2.0. It suits the end of a pipeline: generate and approve a still on an image model, then animate it here for the final clip. Your prompt should talk about movement, camera and sound, because the image already answers what everything looks like.
Ideal use cases
- Animated AI characters. Start from a portrait of a consistent character and give her a gesture, a look to camera, or a walk, with ambient sound.
- Product reveals. Take a clean studio still and add a slow orbit or a light sweep to make a thumb-stopping product clip.
- Turning a hero image into a header video. Convert a strong web image into a short looping-feel clip without re-casting the look.
Limits to know about
- Quality depends on the source. Blurry, cluttered or oddly cropped images tend to warp as they move.
- It takes a single source image, so for several references in one clip you need a reference-capable model such as Gemini Omni Flash 1.1.
- It is the premium tier, so test the motion idea on Fast I2V first.
Example prompts for Seedance 2.0 I2V
Turn and smile
She turns her head from the window toward the camera and gives a small, relaxed smile, curtains moving slightly in a breeze, very slow push-in, soft room tone.
Why it works. One small action with subtle camera movement is the most dependable kind of image-to-video motion.
Product sweep
The camera orbits slowly a quarter turn around the bottle while a soft strip of light sweeps across the glass, quiet studio ambience, no other movement.
Why it works. Restricting movement to the camera and the light protects the product shape from distortion.
Cityscape comes alive
Cars begin to stream along the avenue, clouds drift across the skyline, a flock of birds crosses the frame left to right, distant traffic hum.
Why it works. Environmental motion in a wide establishing image avoids human faces entirely and shows off the model's world animation.
Settings and what costs more
- Length is the main cost driver: 5 seconds costs 104 credits and 15 seconds costs 310 credits.
- Resolution also changes the price: a 5 second clip is 46 credits at 720p and 104 credits at 1080p.
- Start from the aspect ratio of your image so nothing gets cropped. Use 9:16 stills for vertical clips.
- Resolution is selectable on this tier. Use 720p while checking the motion, then render the keeper at 1080p.
- Keep the prompt about movement. Repeating what is in the picture wastes words and can cause drift.
Compare Seedance 2.0 I2V with
- Seedance 2.0 Fast I2V (37 credits per clip)Choose it for cheaper 720p version for testing which stills animate well.
- Kling 3.0 Pro I2V (26 credits per clip)Choose it for accepts a second reference image and offers a different motion feel.
- Seedance 2.0 (104 credits per clip)Choose it for when you have no image yet and want the model to invent the scene.
Our take
Approve the still first, then animate it here. That order protects both your look and your credits.
Use it for
- Building a consistent AI influencer, using reference images to keep one face across content.
- UGC-style product ads, from script to finished clip.
- Choosing a platform? Klingobanana vs Higgsfield, Klingobanana vs OpenArt.
Related models
Frequently asked questions
What should I write in an image-to-video prompt?
- Describe movement, camera and sound only. The image already defines the subject and style, so repeating its description adds nothing and can introduce unwanted changes.
Will the character look the same through the clip?
- Starting from a fixed image gives much better stability than text-to-video. Faces are most reliable with simple movements and a clear, front-facing source.
Does it generate sound?
- Yes, Seedance 2.0 produces native audio. Mention the sounds you expect in the prompt.
Can I use it from the API?
- Yes. Send the image URL, a motion prompt, duration and resolution to the Klingobanana API with this model id, and the video URL is returned when the job finishes.
How many credits does Seedance 2.0 I2V cost?
- A 5 second clip costs 104 credits on Klingobanana. The price table on this page lists every supported length and resolution.
Can I use reference images with Seedance 2.0 I2V?
- Yes, it accepts up to 1 reference image.
Do I need a subscription?
- Yes. There is no free tier; credits come from a subscription plan or a top-up.
Does Seedance 2.0 I2V generate audio?
- Yes, it generates native audio with the clip.
Generate with Seedance 2.0 I2V
Subscribe, open the studio and pick Seedance 2.0 I2V from the model list. Credits are only spent when you generate.
Generate with Seedance 2.0 I2VA subscription is required before generating. Plans from €15/month, cancel anytime.
Default price shown at 1080p where a resolution option exists.