Nine qualified image models. One prompt-to-download workspace.

PHOTO-TO-VIDEO GUIDE · SOURCES CHECKED NOVEMBER 3, 2026

How Do You Make an AI Witch Video From One Photo?

For the exact vertical witchcore reel, use the linked external template and check its current quote before submitting. PromptToVisual does not reproduce that multi-shot template: its current image-to-video form accepts one first frame and one motion prompt for a four-second square or 16:9 clip, so use it only for a simpler custom witch shot.

What the referenced template currently lists

  • One uploaded photo
  • 10–12 second 9:16 template clip with sound
  • The public page listed a quote from 56 credits

Source checked November 3, 2026: MakeViral AI Witch Video template ↗. These are the external page’s current claims, not PromptToVisual test results. Its price, availability and output can change.

Choose a photo that supports the shot

  • Use a clear head-and-shoulders or waist-up photo with one subject.
  • Leave space above the head for a hat and beside the subject for smoke or a prop.
  • Avoid text, other people and important details at the frame edges.

Use a source you have permission to upload. Start with the photo preparation checklist, then compare this format with the other Halloween AI video formats.

Plan the workflow

  1. Choose the exact workflow first

    Open the external template when you want its rotary-phone, cauldron and cat sequence. Use PromptToVisual only when one four-second custom shot is enough.

  2. Prepare a permissibly owned photo

    Crop around one clear subject, but leave room for the requested prop and movement. Keep the original file so you can compare any face or costume changes.

  3. Write one short shot

    For a custom generation, request one action and one camera move rather than the external template’s full multi-shot story. Review the displayed model, aspect and paid-credit quote before submitting.

  4. Inspect before sharing

    Check the face, hands, hat edges, smoke, unwanted lettering and audio. A completed generation is not proof that the person or scene stayed exact.

Four-second custom prompt example

The subject lifts a black witch hat slightly and looks toward a softly glowing cauldron. Warm smoke curls upward; the camera moves slowly closer; subtle room ambience. Keep one subject and one continuous shot.

This is an original writing example, not a tested prompt or a promise of a particular result. Review the current studio quote before any submission; preparing or copying the prompt does not generate a video.

Check the finished clip

  • One recognizable subject remains in frame.
  • The movement fits a single four-second shot.
  • No extra face, hand, text or logo appears.
  • Any music or source photo is cleared for the intended use.

Common questions

Can PromptToVisual copy the full witchcore reel?

No. The general form offers one first frame, one prompt and a four-second square or 16:9 result. It does not provide the external template’s vertical multi-shot sequence or original audio.

Was this custom prompt tested?

No. It is a writing example only. No new video was generated for this guide.

Can I use someone else’s photo?

Use a photo you have permission to submit and publish. A public photo is not automatically cleared for a face-replacement video.

Choose the next step

Open the exact external template ↗

Open the shorter PromptToVisual video form · Compare the three formats · Prepare the photo

No hands-on output testing was performed for this page.

Checking image studio availability…