AI image animation model benchmark
Seedance 2 team · Updated
We animated the same AI portrait with five image-to-video models on Seedance 2, using one motion prompt. In the frames we checked, all five kept her face, hair, sweater and scarf from the photo. Veo 3.1 cost least (10 credits for 8 seconds); Seedance 2.5 cost 60 credits for 5 seconds at 720p and can also make the person speak a quoted line, with the voice generated in the same pass. Watch the clips to compare the motion.
The photo and the prompt
Start frame: a portrait of a fictional woman made with Nano Banana 2.1 on Seedance 2 for 1 credit (https://r2.seedance.tv/images/1d599ce5-cdb6-48de-88a7-85ef3b8e8b07-1794488022505-0.jpg).
She turns her head toward the camera, smiles and tucks her hair behind her ear as a light wind lifts her scarf; slow push-in. Sound: light breeze, distant city.
Settings: image to video, 5 seconds at 720p (Veo 3.1 only makes 8-second clips), sound on.
Results
Times include queueing; all five jobs were sent at once. Veo's time was lost to a network error while we polled it; the clip itself finished.
| Model | Clip | Credits | Time to video |
|---|---|---|---|
| Seedance 2.5 | 5 s, 720p | 60 | 4 min 47 s |
| Seedance 2.0 | 5 s, 720p | 39 | 2 min 27 s |
| Kling 4.0 | 5 s, 720p | 35 | 3 min 56 s |
| Veo 3.1 | 8 s, 720p | 10 | not recorded |
| Wan 3.0 | 5 s, 720p | 20 | 2 min 16 s |
Watch the clips
Tips for animating a photo
- Use a sharp, well-lit photo; the model keeps what it can see.
- Describe one clear action and one camera move, as in the prompt above.
- Only animate photos of yourself or people who agreed to it.
- To make the person speak, put the line in quotes with sound on; Seedance 2.5 makes the voice and moves the lips to match.
How this was made
Made on 2026-11-12 through the Seedance 2 MCP server (https://www.seedance2.net/mcp), the same way Claude, ChatGPT or Codex would call it, on a paid Seedance 2 account, so there is no watermark. Credits and times are what each run actually charged and took.
Try the prompt
Portrait, wind and a smile
Image to videoUpload a portrait as the first frame.
She turns her head toward the camera, smiles and tucks her hair behind her ear as a light wind lifts her scarf; slow push-in. Sound: light breeze, distant city.
Questions
- Which AI model is best for animating a photo?
- All five here kept the person recognisable from one photo. Choose Seedance 2.5 when the person should speak or you need several reference images, and Veo 3.1 for the cheapest 8-second take.
- Can I animate an old family photo?
- Yes, with image to video on any of these models. Results are best with a clear, sharp scan.