Muse Video

Muse Video, from Meta Superintelligence Labs: 5 to 10 second 720p clips with native sound, made from a text prompt.

Prices

A clip costs its length times the rate: a 5 second 720p clip is $0.10. It’s charged only when the video is delivered.

What it takes

• A prompt of up to 8,000 characters.
• Text only: no first_frame or reference_images. Describe the scene, the camera and the sound in the prompt.
• Lengths from 5 to 10 seconds, in whole seconds.
• 16:9 and 9:16. A 16:9 clip is 1280 × 720 and a 9:16 clip 720 × 1280.
• Sound is always on.
• 720p.

Example

{
  "model": "muse-video",
  "prompt": "A street musician plays violin on a rainy Paris evening, warm shop lights reflecting on wet cobblestones. We hear the violin melody and the rain.",
  "seconds": 10,
  "aspect_ratio": "9:16"
}

Prompting

• Say what happens, where, and how the camera moves, then what we hear: ambience, effects, music or a line in quotes.
• Keep one clear event per clip; Meta says sound-to-picture sync and fast motion are still improving.
• Check dialogue clips for lip sync before you use them.

Good to know

• Clips usually take one to three minutes.
• It makes video from text only; to start from an image, use Wan, Kling or Veo.
• Meta hasn’t published its own API for Muse Video; on Zurelay it works like every other video model.
• Prompts or images that may break the content rules can be declined (content_policy); declined videos are free.
• Videos are kept for 30 days; download the ones you want to keep.

Full API details: Video generation. Try it in the dashboard’s Playground, which shows the code for each video.