Muse API

October 12, 2026 (October 30, 2026)

This is a third-party API for Muse by Meta. It uses your Muse subscription: Free, Power $20 or Maximum $100 a month (as of October 2026). Through a REST API and a hosted MCP server, it makes images and 10-second videos with sound and spoken dialogue, and lets you chat with the Muse agent and watch or take over its web browser. Meta sells no Muse video API: its own Model API has no video model and no API for the Muse agent (as of October 2026).

What it makes

Output	What you get	Notes
Image	WebP, about 2 to 2.5 megapixels	From text or up to 9 of your images: combine photos, change an outfit, restore and colorize an old photo
Video	MP4, about 10 seconds at 720p, with sound	From text, a photo or another video, with up to 9 images, 3 videos and 3 audio files. Quoted lines are spoken, so a photo talks, and a person in a clip can be swapped for someone else
Chat	Text, files and cards, streamed live	Research, code, connected apps, documents, web apps and podcasts, with your images, videos, audio and workspace files attached
Library	Every file on the account	Uploads up to 50 MB, made media, documents (also as PDF), podcasts as MP3, and web apps you can publish on a public link
Workspace files	Any file on the agent’s computer	Search, download, write and delete the code, documents and data the agent writes

Muse picks its own model, so there is no model, resolution or duration setting: you ask for the aspect ratio and the spoken lines in the prompt, and Muse follows any common ratio. Every file Muse makes is an asset id that the next job takes as an input, so edits chain with no downloads or re-uploads. In our tests an image took 20 to 60 seconds and a video 60 to 140.

Agent

Chat with Muse, Meta’s personal AI agent, through POST /chats: its answers stream live and pick up again after a dropped connection, and you can send the next message mid-answer, stop it, and list, read or delete chats. When it opens a website, watch its browser live and take control with mouse and keyboard. Answer its permission requests and quick-reply cards, set what it may do without asking, or reset a stuck account to a fresh agent computer. The live demo is a complete chat client built on it, and we keep adding more of what Muse can do.

MCP server

The hosted Muse MCP server gives Claude Code, Cursor, VS Code and other AI assistants 37 tools for Muse images, videos, chat, the agent’s browser and the Library, with your useapi.net API token and nothing to install. The muse-api repo also adds it in one step as a Claude Code plugin or a Gemini CLI extension.

Pricing

How the useapi.net Muse API compares with Meta’s own API (October 2026). Muse has no credits and no price per image or video: each job uses part of your Muse account’s weekly allowance, and a paid Muse plan gets a bigger one.

	useapi.net Muse API	Meta’s Model API
Cost	Your Muse subscription, the free plan included	$0.01 per image
Images
10-second videos with sound and speech
Chat with the Muse agent at muse.ai
The agent’s live web browser, Library and files

A free Muse account makes images, videos and chats, and each account you connect has its own weekly allowance.
Meta publishes each plan’s weekly allowance in Muse tokens: 500 million on Power and 3 billion on Maximum, while the free plan’s limit isn’t published (About Muse subscriptions). Meta publishes no token cost per image or video. In our tests on the free plan, a video used 2 to 3 percent of the week and an image 0 to 1 percent.
GET /accounts/email shows each account’s plan, the share of the week used and when it resets.

Start using the API

A single $15/month subscription covers every API on useapi.net — one token, all services. You connect up to 3 accounts per service, each with its own website subscription, and useapi.net handles the REST API, load balancing, and polling. Need more accounts? Each extra subscription is $10/month and adds 3 more accounts per service.

Subscribe — $15/month Pay with crypto

Pricing and volume options — full refund within the first 14 days if you have fewer than 50 successful generations. Already subscribed? Change the quantity in the Stripe Customer Portal instead of buying again.

Run several Muse accounts

Each useapi.net subscription adds 3 Muse accounts. Image and video jobs don’t need email: a job goes to the account that holds its inputs, or to a healthy account with allowance left and a free slot within its maxJobs (1 to 50, default 10). An account whose weekly allowance is used up sits out until it resets, and a job sent to it with email answers 412. An account that reached Muse’s video limit for the last 24 hours sits out video jobs for an hour. When every account is busy you get 429, so wait and send it again. GET /jobs shows what each account is running, and chats take no job slot.

Every job can post its finished record to your webhook with replyUrl, or you poll GET /jobs/jobid, so a video needs no open connection.

How useapi.net works has tips for keeping your accounts healthy.

Questions

Is there a Meta Muse API?

Through useapi.net, yes, for video, images and the Muse agent: the useapi.net Muse API makes 10-second videos with sound and spoken dialogue, images and live chat with the agent, on your Muse subscription. Meta’s official Model API sells the Muse Image model at $0.01 per image, but as of October 2026 it has no video generation and no API for the Muse agent at muse.ai, where the videos are made.

How much does the Muse API cost?

Through useapi.net, an image or a video has no price of its own: it uses the weekly allowance of your Muse subscription, and the free plan works. A paid Muse plan raises the weekly allowance, as the comparison shows.

Is there a Muse API key?

Through useapi.net, one API token is all the key you need, for the REST API and the MCP server alike. Add your muse.ai email and password once with Setup Muse, and we keep the account signed in, with no cookies to copy. Create the account in the US, Canada, Japan or Singapore, where Muse is available (as of October 2026), and finish its onboarding at muse.ai. After that you can call the API from anywhere.

Setup Muse
LLM-friendly API spec Feed this to your LLM to build integrations
Claude Code skill A drop-in useapi.net skill for Claude Code, and a rule for Cursor
Postman collection (October 14, 2026)
MCP Server Muse images, videos, chat and the agent’s browser from Claude Code, Cursor, VS Code and other MCP clients
Chat Streaming Guide Chat with the Muse agent, its answers streamed live over SSE
Live Browser Guide Watch the agent’s web browser live and take control of it
Live Demo Try it in your browser. Source code on GitHub
Agent skill: muse-video Claude Code and Codex make Muse images and videos for you: npx skills add useapi/muse-api --skill muse-video
GitHub examples Node.js and Python: combine photos, restore old photos, make a photo talk, swap a person in a video and make a narrated 30-second Short, plus the live chat demo, n8n workflows and an agent skill
n8n template: Old Photo to Talking Video Restore an old photo, then make the person in it speak your line, with pick-or-redo pages and an optional Google Sheet batch mode, ready to import
n8n template: Narrated Faceless Shorts Type a topic: Muse writes a short script and makes one narrated 9:16 scene per part, joined into one Short and saved to Google Drive