Slatecraft

Vidu Q4 reference to video

Give Vidu Q4 Preview the cast and the props (up to 12 images of characters, outfits, products and places, plus up to 3 voice clips) and describe the scene. The model builds a new shot around them, keeps every character recognizable across camera cuts, and matches their voices.

Characters, outfits, products or places. In the prompt, @1 is the first image, @2 the second, and so on.

MP3, 3–12 seconds each. Keeps a character's voice consistent.

Length

05sec

3s16s

Resolution

Aspect ratio

Sound

The Duel

Made on Slatecraft, Reference to Video · 720p · 10 s, with sound

Reference images

  • The Duel: @1@1
  • The Duel: @2@2
  • The Duel: @3@3

EXT. STONE BRIDGE - RAIN

In @3, @1 and @2 face each other in the middle of the bridge in the pouring rain, lanterns swaying. @2 raises his sword and says: "Only one of us leaves this bridge." @1 draws his katana. They charge and their blades clash in a burst of sparks. Cut to a slow-motion close-up of the crossed blades with rain bouncing off the steel. Thunder, rain, the ring of steel on steel.

How it works

  1. 1

    Add your references

    1 to 12 images, one subject each works best: a face, an outfit, a product, a location. They are numbered @1, @2… in the order you add them.

  2. 2

    Write the scene

    Use the numbers in the prompt: “@1 hands @2 the letter from @3.” Add up to three MP3 voice clips (3–12 s) to keep voices consistent.

  3. 3

    Frame and generate

    Pick 16:9, 9:16, 1:1, 4:3 or 3:4, a length from 3 to 16 seconds and a resolution up to 4K.

Reference to video prompts that work

  • Name every reference

    Each image should appear in the prompt by its number. An image you never mention may be ignored or blended into the scene.

    @1 sits on the sofa from @3, reading the book from @2. She looks up and laughs. Slow push-in, evening lamp light.

  • Short drama: write the beats

    Vidu Q4 can cut between angles on its own. Give it the beats in order (action, line, reaction) and it plans the coverage.

    @1 and @2 face each other in a dark corridor. @1 says “You knew all along.” Cut to @2, silent, then a close-up of @1.

  • Ads: product plus presenter

    Put the presenter in one image and the product in another, then describe how the product is shown.

    @1 holds the phone from @2 up to the camera and says “Two days of battery, finally.” Bright studio, quick zoom on the screen.

  • Keep a voice across videos

    Reuse the same MP3 for a character in every clip of a series. Clips of 5–10 seconds of clean speech work best.

Vidu Q4 reference to video FAQ

  • From 1 to 12 images per video, plus up to 3 reference sounds. JPG, PNG or WebP up to 20 MB each.

  • By its number: @1 is the first image you added, @2 the second, and so on. The numbers are shown on each thumbnail in the editor.

  • They guide how characters sound: tone, accent and delivery. Each clip must be an MP3 of 3 to 12 seconds. Keep Sound on so the video is generated with dialogue.

  • No. Both modes cost the same per second at each resolution; images and voice clips are free to add.

  • Only with their permission, and never to make sexual, hateful or deceptive content about them. Public figures are often blocked by moderation; blocked videos are refunded.