Short video or motion graphic from text or image, free option

Goal

Produce a 15–30 second vertical clip (Reels/TikTok/Shorts) from an idea, using free tiers first.

Steps

1. Use the script prompt below in ChatGPT to get a 4–6 shot script with on-screen text and one visual prompt per shot.
2. Generate each shot (5–8 s) in Google Flow (Veo) from the visual prompt, or animate a still image (image-to-video). Keep the same style words in every prompt.
3. For text-heavy motion graphics (titles, numbers, logos), use Canva's animated templates instead of a video model.
4. Import the shots into CapCut, put them in order, add the on-screen text and auto-captions, and trim to the beat.
5. Add music from the app's licensed library, export 1080×1920.

Prompt

Replace anything in { } with your details.

Write a {LENGTH}-second vertical video script about: {TOPIC}. Audience: {AUDIENCE}. Language of on-screen text: {LANGUAGE}.
Return a table with 4–6 shots. For each shot give: duration (s), on-screen text (max 7 words), voiceover line, and a visual prompt in English for a video model that includes subject, action, camera move, lighting and the style words "{STYLE}" (repeat the same style words in every shot).
Shot 1 must hook in under 2 seconds. End with a clear call to action.

Expected output

A captioned 15–30 s vertical video ready to post, plus the script and prompts to remake it.

Gotchas

• ⚠ Free tiers have daily or monthly credit limits and may add watermarks; check before you plan a big batch.
• ⚠ Video models still garble on-screen text, especially Arabic: add text in CapCut/Canva, not in the prompt.
• ⚠ Don't generate real people or brands you don't have rights to.
• ⚠ Availability of each generator is unverified.