GPT IMAGE 2.5 | WAN 3.0 | SEEDANCE 2.5 IS NOW LIVE!

GENERATE YOUR VIDEOS WITH UNLIMITED SEEDANCE 2.5 TODAY.

NO EXTRA SUBSCRIPTION
REVIRAL
Get API key
Browse API docs

Endpoints

AI Video Prompt Builder API

Was this page helpful?

Use the free AI video prompt builder API to convert a creative brief and optional reference images into structured, moderated production instructions.

Keyword target: AI video prompt builder API

The prompt-builder endpoint converts a campaign brief into structured instructions for a video workflow. It accepts timing, style, effect settings, and up to four embedded reference images. Prompt writing is free; a render is charged only after the resulting prompt is submitted to generation.

  1. Describe the subject, action, camera, setting, and intended outcome in brief.
  2. Set the duration and style fields; include images only when the prompt must describe supplied visual references.
  3. Review the generated prompt and submit it to generate-media or Director chat for final planning.

Working example

Set REVIRAL_API_KEY only in your server environment, then run this request with values valid for your account.

curl -X POST "https://reviral.ai/api/v1/prompt-builder" \
  -H "Authorization: Bearer $REVIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  --data '{"brief":"A clean vertical product reveal for a leakproof travel mug during a morning commute.","durationSec":15,"section1MaxChars":1000,"styleId":"cinematic-product","effectEnabled":false,"effectId":null,"images":[],"compact":true}'
Example response
{
  "data": {
    "ok": true,
    "output": {
      "section1Timeline": "Vertical cinematic product reveal, morning train platform, tactile lid close-up, natural handheld movement, clean final product frame."
    },
    "variationSeed": "mug-launch-01"
  }
}

Frequently asked questions

Does the prompt builder start a video render?
No. It returns text only. Submit the approved prompt to a generation endpoint to start a priced media job.
How many reference images can I include?
You can include up to four supported image payloads, each within the documented encoded-size limit.
When should I use Director after prompt builder?
Use Director when a user wants a conversation that critiques or refines the generated production prompt before rendering.
POST

/api/v1/prompt-builder

Build a video prompt

Uses the same moderated free structured prompt writer as the product.

Authentication: Bearer API key required

curl -X POST "https://reviral.ai/api/v1/prompt-builder" \
  -H "Authorization: Bearer $REVIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  --data '{"brief":"A clean vertical product reveal for a leakproof travel mug during a morning commute.","durationSec":15,"section1MaxChars":1000,"styleId":"cinematic-product","effectEnabled":false,"effectId":null,"images":[],"compact":true}'

Request body

FieldTypeRequiredDescription
briefstringYes
durationSecinteger (min 5, max 300)Yes
section1MaxCharsvalue (min 500, max 3500)Yes
styleIdstringYes
effectEnabledbooleanYes
effectIdvalueYes
imagesarrayYes
variationSeedstringNo
compactbooleanNo

Responses

200
Prompt response
400
Settings are invalid or a reference field is unsupported
401
Key is missing, revoked, or invalid
422
Settings are invalid or a reference field is unsupported
429
Too many requests
502
Job could not be started
503
Models or rate limiting are temporarily unavailable