Muse Video — 10-second clips with sound from a prompt or your images
Muse Video is Meta's video model: describe a shot, or add up to 9 reference images, and get a 10-second 720p clip with sound. $0.024 per clip, nothing to subscribe to.
- Text to video
- Up to 9 references
- Sound in every clip
- 10 seconds, 720p
- Seven aspect ratios
- Pay per clip
Try Muse Video before you write any code
Run Muse Video with the exact request your code will send. Every setting that changes the clip is here.
muse-video$0.024 per clip
Describe the shot, add reference images if you have them, pick a shape, and submit once. The Muse Video clip arrives with the exact amount it cost.
Subject, action, camera move, light and the sounds you want. Muse Video needs a prompt even with references.
Files are uploaded and sent as URLs.
Optional. Up to 9 images, 11 MB in total. They guide the clip; the first frame is not an exact copy.
Auto follows your references; a text-only clip comes back 16:9.
One price per clip at current rates.
Try one of these prompts
Generated with Muse Video on SeedRouter
Muse Video pricing
Muse Video has one price per clip. Every clip is about 10 seconds at 720p with sound, so text-only and reference requests cost the same.
- Muse Video
- $0.024/clip
- Muse Video
- 833clips
Muse Video prices come from the rates that bill your account. Failed requests are not charged.
- Muse Video
- $0.024Price · 1 clip
What is Muse Video?
Muse Video is the video model from Meta Superintelligence Labs, previewed on July 7, 2026. Meta built it on the same base as Muse Image and gave it native audio, so the picture and the sound come out together.
On SeedRouter, one Muse Video request takes a prompt, up to 9 reference images and an optional aspect ratio, and returns one clip of about 10 seconds at 720p. You send the request once, keep the task ID, and read the video URL when the task completes.
In our tests on October 29, 2026, Muse Video clips came back as 10.0-second MP4 (H.264) files with an AAC audio track: 1280×720 at 16:9, 720×1280 at 9:16, in about three minutes each.
Two ways to use Muse Video
Call the API from your backend, or hand the job to a coding agent.
Call the API
Send the Muse Video request from your server, keep the task ID, and read the finished clip from the task.
- 1Create an API key
- 2POST the request with your prompt
- 3Poll the task until the video is ready
Let an agent do it
Copy a ready Muse Video prompt into Claude Code, Codex or Cursor. It asks before spending and downloads the result.
- 1Copy the prompt
- 2Paste it into your agent and describe the shot
- 3Approve the cost, then get the video
What Muse Video is good at
Muse Video fits work where a short brief should come back as a finished clip with sound.
Sound in every clip
Rain, wind, crowds or music: Muse Video renders the audio with the picture, so there is nothing to add later.
Start from a prompt
Describe the subject, the action and the camera move, and Muse Video returns a full 10-second shot.
Or from your images
Give Muse Video up to 9 references, such as a product, a character or a still from Muse Image.
Seven shapes
Vertical for Reels and TikTok, wide for YouTube, square for feeds: one field changes the shape.
Camera moves
Ask for a push-in, a tracking shot or a slow drift, and the camera moves to match.
One flat price
Every Muse Video clip costs the same, so a batch of drafts is simple to budget.
What teams use Muse Video for
Social clips
Vertical clips for Reels and TikTok, with sound already in, ready to post.
Product shots
Turn a product photo into a short Muse Video shot with motion and sound, then test a few prompts.
Stills that move
Make a picture with Muse Image, then pass it to Muse Video as a reference and set it in motion.
B-roll with ambience
Generate nature, city and studio clips with background sound for edits, explainers and presentations.
Storyboards
Turn a scene description into a 10-second draft before planning a real shoot.
In-app video
Give your users a text-to-video or photo-to-video button backed by the Muse Video API.
Start with Muse Video in four steps
- 01
Create a key
Sign in, top up any amount, and create an API key. Credits never expire.
- 02
Send one request
POST your prompt to /v1/videos/generations with model muse-video, plus image_urls and aspect_ratio if you need them.
- 03
Poll the task
Keep the Muse Video task ID and check it every 10–20 seconds until the status is completed.
- 04
Download the video
Save the hosted Muse Video URL from the task to your own storage.
Muse Video at a glance
The facts that decide whether Muse Video fits the job.
- Clip length
- About 10 seconds
- Resolution
- 720p
- Inputs
- Text, up to 9 reference images
- Audio
- Sound in every clip
- Aspect ratios
- Seven, or auto
- Billing
- One price per clip
Need 1080p, a chosen length, first and last frames or silent video? See Kling 4.0.
What to know before you start
A prompt is always required
Even with references, Muse Video needs a prompt that says what happens in the shot.
References are sent as URLs
Send each reference as a public HTTP(S) link, 11 MB in total, not base64. The Playground uploads files for you.
Fixed length and resolution
Every Muse Video clip is about 10 seconds at 720p with sound; there is no switch for these.
Results arrive by task
A Muse Video request returns a task ID first. Poll it until the video is ready, about three minutes in our tests.
Why run Muse Video on SeedRouter
Pay per clip
Muse Video is billed per completed clip at the rate shown above. No subscription and no monthly minimum.
Failed requests are free
If a Muse Video task ends failed, nothing is charged. You only pay for videos that are delivered.
Credits never expire
Top up once and use the balance whenever you need it.
Hosted results
Finished clips come back as URLs on our CDN. Keep a copy of anything you need long term.
One balance for every model
The same top-up pays for Muse Video and every other video, image and text model in the catalogue.
Other video models on SeedRouter
Same balance, same task-based workflow.
Muse Video guides
Practical guides to Muse Video pricing, prompting and comparisons.
Muse Video FAQ
What is Muse Video?+
Muse Video is the video model from Meta Superintelligence Labs, previewed in July 2026 alongside Muse Image. On SeedRouter it makes one clip of about 10 seconds at 720p with sound, from a prompt or from a prompt with up to 9 reference images.
How much does Muse Video cost?+
Muse Video costs $0.024 per clip on SeedRouter, sound included. Every clip is about 10 seconds, so the price is not multiplied by seconds, and failed requests are not charged.
Is Muse Video free?+
Not on SeedRouter. The Muse Video API is paid per clip, and new accounts get a small credit to try it. There is no subscription: you top up any amount, credits never expire, and failed requests are not charged.
How do I use the Muse Video API?+
Create a SeedRouter key and POST your prompt to /v1/videos/generations with model muse-video. You get a task ID back; poll /v1/tasks/{id} until the video is ready, then download it from output.video_url.
Does Muse Video generate sound?+
Yes, always. Every Muse Video clip has a sound track made together with the picture, and audio: false is refused. Name the sounds you want in the prompt.
Can Muse Video animate my own image?+
Yes. Pass up to 9 image URLs in image_urls and describe the motion in the prompt. Muse Video uses them as references; the first frame is close to your image but not a pixel copy.
Can I make a longer Muse Video clip or pick 1080p?+
No. Muse Video makes one clip of about 10 seconds at 720p; duration and resolution accept only those values. For longer or 1080p clips, Kling 4.0 is on the same balance.
Can Muse Video make vertical video?+
Yes. Set aspect_ratio to 9:16 for a vertical clip; 16:9, 1:1, 4:3, 3:4, 3:2 and 2:3 also work. Leave it out and Muse Video follows your references, or 16:9 for text only.
Generate your first Muse Video clip
Try Muse Video in the Playground, then move the same request into your code.