Vidu Q4 API

Vidu Q4 (Oct 2026) from ShengShu — animates a start image into a 720p clip of 3–16 seconds with native audio. Image-to-video only. $0.0665/s at fal's list price.

$0.0665/s on Varosity credits ($0.0665/s with your own key). Clips up to 16s with native audio.

Use Vidu Q4 with your agent Try it in the browser

no falaccount needed · provider's list price, no markup · failed renders never billed

Best for

native audio
3–16s clips
low price

Trade-offs

needs a reference image (image-to-video only)
audio can't be turned off
720p on Credits

How to get the best result

Low-cost image-to-video: animate a start frame into a 720p clip of 3 to 16 seconds with native audio. Pick instead: A text-to-video model when you have no image (grok-imagine-video or veo-3.1-lite for cheap clips with audio); veo-3.1 for lip-synced dialogue; a silent model (e.g. seedance-1.0-pro) for a clip you will score yourself.

Prompting

Describe the motion and the camera move, not the picture: the image already sets subject, framing and style.
Describe the sound you want in its own sentence (ambience, effects, music); audio is made with the picture.

Settings on Varosity

Needs referenceImageUrl (the start frame); a request without one is refused before anything is charged.
durationSec is whole seconds from 3 to 16, billed per second produced at $0.0665/s.
Always has audio (no off switch). aspectRatio must be 16:9, 9:16 or 1:1, but the clip follows the image's framing.
720p on Credits; providerOptions.resolution (540p, 1080p, 2K, 4K) only with your own fal key.

Avoid

Calling it without an image; it is image-to-video only.
Expecting a silent clip; it always comes with generated sound.
Negative prompts; Vidu Q4 has no negative-prompt field.

Example prompt

The boat rocks gently on the swell, gulls circle and call overhead, water laps against the pier.

From fal.ai, checked against what Varosity sends.

Call it from your code

One Varosity key works for every model. Video renders are async — the response has a jobId; poll GET /api/v1/jobs/{jobId} until status is "succeeded", then read outputUrl.

Before you call it

vidu-q4 needs a reference image (referenceImageUrl).

curl -X POST https://varosity.ai/api/v1/video/generate \
  -H "Authorization: Bearer $VAROSITY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "cinematic aerial of a coastal road at golden hour, slow push-in",
    "modelId": "vidu-q4",
    "aspectRatio": "16:9",
    "durationSec": 5
  }'

Or ask your agent

Connect Claude, Cursor, or any MCP client to https://varosity.ai/api/mcp with your key, then say:

"Use Varosity with vidu-q4 to make a 5-second clip of …"

Agent setup guide