AI Video Generator
Reference to Video
Create video from ordered image, video, and audio references. Choose Seedance 2.5 or MiniMax H3.
Reference to video features
Ordered multi-reference input
Add up to 30 images, 10 videos, and 10 audio clips in one ordered reference set. The workspace numbers each item so you can call out Image 1, Video 2, or Audio 1 in the prompt.
Camera motion from reference video
Use one of the 10 video slots to provide a camera move for the output to follow. Name the numbered clip in the prompt so the motion cue points to a specific reference.
Audio as a first-class reference
Use up to 10 audio slots for voice, rhythm, or sound references. MiniMax H3 also accepts an audio file as the only reference, so that route does not require an image.
Two current R2V models
Choose between Seedance 2.5 and MiniMax H3, the two reference to video models currently available in the workspace. The model selector stays on the same page, so you can switch models without rebuilding the reference set.
Models for Reference to Video
2 models support this task type. Open any one for full specs.
Generate AI video with Seedance 2.5: text to video, image to video, and reference to video, all from one Vidney workspace.
Generate AI video with MiniMax H3: text to video, image to video, and reference to video, all from one Vidney workspace.
How to create a reference to video clip
Create a reference to video clip in four steps using the two models available in the Vidney workspace.
Frequently asked questions
What is the Vidney Reference to Video?
The Vidney Reference to Video lets you create AI video from reference to video using any supported model in one workspace.
Which models can I use with the Reference to Video?
You can use Seedance 2.5 and MiniMax H3.
Do I need an API key to use the Reference to Video?
No. Vidney handles provider access and billing, so there are no API keys to manage and no provider account to set up.
