Models
AI Video Models
Three models behind one prompt box. Here is what each one does, side by side, so you can pick the right one for the clip.
Side by side
The limits below are the ones the generator uses. If a setting is not listed for a model, that model does not offer it here.
| 2.5Seedance 2.5 | 2.0Seedance 2.0 | H3MiniMax H3 | |
|---|---|---|---|
| Made by | ByteDance | ByteDance | MiniMax |
| Length | 4–30s | 4–15s | 4–15s |
| Quality | 480p · 720p | 480p · 720p | 2K |
| Shapes, text to video | 9:1616:91:13:44:321:9 | 9:1616:91:13:44:321:9 | 9:1616:91:13:44:321:9 |
| Shapes, image to video | Follows your image | 9:1616:91:13:44:321:9 | Follows your image |
| Sound | Sound, can be muted | Sound, can be muted | Sound always on |
Start from words or from a picture
FAQ
Choosing a model
Which model is right for my clip?
Go by the clip. Over 15 seconds: Seedance 2.5. A 2K picture with sound: MiniMax H3. A silent clip, or a fixed output shape from your own image: Seedance 2.0. For short social clips any of the three works, so try the one whose look you like.
Can I switch models after writing a prompt?
Yes. The model is a button in the bar under the prompt. Switching keeps your prompt and image and moves the length and quality to the nearest ones the new model offers.
Do all models make sound?
Yes, all three generate audio with the picture. Seedance 2.5 and 2.0 let you mute it; MiniMax H3 always renders sound.
Do the models cost the same?
No. The price depends on the model, the length and the quality, and the Create button shows it before you pay. Whatever you pick, credits come back if a render fails.