AI Generator

Reference to Video AI

Mix images, video clips and audio — tag each one in your prompt with @ and turn them into one video.

Reference to Video AI

Give the AI your character art, a motion clip and a soundtrack, then say how they fit together. Tag each reference with @ in one prompt and get a single video where your character looks right, moves right and sounds right.

reference to video AI: character sheet, motion clip and audio combined into one anime video

What You Can Do with Reference to Video

  • 01

    Keep your OC on-model in every shot

    Upload front, side and outfit sheets of the same character. Every image you tag becomes part of the identity, so the face, hair and costume stay consistent instead of drifting the way a single text prompt does.

  • 02

    Borrow motion and camera work from a clip

    Add a short reference video and write “move like @Video 1”. The dance, the walk cycle or the camera push-in carries over to your character without describing it frame by frame.

  • 03

    Let a voice or a beat drive the scene

    Attach a voice line, a song hook or a sound effect as audio. The video is generated with that sound in mind, so movement and rhythm follow it.

  • 04

    Put characters, props and places together

    Tag a character, a prop and a background separately: “@Image 1 sits on the chair from @Image 2 in the room from @Image 3”. One prompt, one coherent scene.

How to Make a Video from References

  1. Add your references

    Upload up to 10 images, 5 video clips and 5 audio clips. Reference videos can be 15 seconds in total, and so can audio.

  2. Tag them in your prompt

    Type @ to pick a reference and describe what each one should do, for example: “@Image 1 dances like @Video 1 to @Audio 1 on a neon rooftop.”

  3. Choose length and quality, then generate

    Pick 480p, 720p or 1080p, an aspect ratio (or Adaptive to follow your references) and 2–30 seconds. Reference video length plus output length must stay within 30 seconds.

What Reference to Video Accepts

  • 01

    Up to 10 images

    Character sheets, outfits, props, backgrounds or style references, tagged as Image 1 to Image 10.

  • 02

    Up to 5 video clips

    Motion or camera references, 15 seconds in total. Their length plus the output length must stay within 30 seconds.

  • 03

    Up to 5 audio clips

    Voice lines, music or sound effects, 15 seconds in total.

  • 04

    2 to 30 seconds of output

    Choose any whole number of seconds — a 10-second reference clip leaves room for up to 20 seconds of output.

  • 05

    480p, 720p or 1080p

    Higher resolution costs more credits per second; the exact cost is shown on the Generate button.

  • 06

    Five aspect ratios plus Adaptive

    16:9, 4:3, 1:1, 3:4, 9:16, or Adaptive to follow the shape of your references. Sound is generated with the video by default.

You may want to know

What is reference to video?

It is a way to make an AI video from several pieces of media at once. Instead of one start image, you give the model a set of references — character images, motion clips, audio — and tell it in the prompt how they relate. The model uses all of them together to generate one clip.

How is it different from image to video?

Image to video animates one picture as the first frame. Reference to video does not lock the first frame; it treats every image, clip and sound as a reference, so you can combine a character from one image with a setting from another and motion from a video.

How do I refer to a specific reference in my prompt?

Type @ in the prompt box and choose the reference. It appears as a tag such as Image 1, Video 1 or Audio 1, numbered in the order you added each type. Removing a reference renumbers the tags for you.

How long can my video be?

Between 2 and 30 seconds. If you add reference videos, their total length plus the output length must stay within 30 seconds — for example, a 10-second reference clip allows up to 20 seconds of output.

Which model powers it?

Wan 3.0 (Reference). It accepts image, video and audio references in one request and can generate sound along with the video.

Does it keep my character consistent?

Consistency is the main reason to use it. Several clear images of the same character from different angles work better than one; avoid mixing two different characters in the same image.

Do I need a membership?

Yes. Reference to Video is available to members, and the cost depends on resolution and length — the exact number of credits is shown on the Generate button before you start.

Bring Your Characters, Motion and Sound Together

Add your references, tag them in one prompt and turn them into a single video with OCMaker AI.

Reference to Video AI: Images, Video & Audio to One Clip | OCMaker AI