The MakeAIVideo API
Generate a finished narrated video from a prompt or a script with a single call. 65 REST endpoints, an MCP server, agent integration and a CLI, all on the same API key and the same credit allowance as the app.
Use cases
What you can build with the AI video API
Video from your own content pipeline
POST a script or a prompt from your CMS, your database or a cron job, and get a finished narrated MP4 back. No editor in the loop.
Videos endpointsA recurring channel that runs itself
Generate a video per row, per article or per product on a schedule, then push the result wherever you publish.
QuickstartA consistent on-screen character
Save a character once and reuse it across every render so an AI presenter stays the same person from video to video.
Characters endpointsPost the finished video for you
Post a finished video now, or schedule it, to connected TikTok, Instagram, YouTube, Facebook Pages, LinkedIn, X, Threads, Pinterest, Bluesky, Telegram and Discord accounts. X takes videos up to 140 seconds.
Publishing endpointsRender completion in your own app
Subscribe to a webhook and get called when a render finishes or fails, instead of polling for it.
WebhooksOne key
Five ways to call the video generation API
One API key works across all of them.
REST API
65 endpoints across 13 groups: videos, characters, scripts, voices, uploads, brand kit, webhooks and more.
Browse the API referenceMCP server
Connect Claude, or any MCP client, straight to your account so an assistant can generate video as a tool call rather than by pasting an API key into a script.
Set up MCPAgents
Let an assistant generate finished video as a tool call, with free cost estimates before it spends and idempotent retries when it stumbles.
Build with agentsCLI
Drive the same API from a terminal or a CI job, for scripted and batch rendering, with no secret pasted into a shell.
Use the CLITypeScript SDK
A typed client that handles polling, webhook signature verification and error classes, so the common path is three calls rather than a fetch loop.
Use the SDKGetting started
Start with the video generation API in three steps
- Step 1
Create an API key
In your account settings. The same key authenticates the REST API, the CLI and the MCP server.
Authentication - Step 2
Make your first render
POST a prompt or a script and get back a video id to track.
Quickstart - Step 3
Get told when it is done
Subscribe to a webhook rather than polling, and handle the finished MP4 in your own app.
Webhooks
Open source
Open source AI video API packages
MIT licensed, on GitHub at github.com/makeaivideo-ai and on npm under @makeaivideo. The OpenAPI spec is at app.makeaivideo.ai/api/v1/openapi.json and the hosted MCP server at https://mcp.makeaivideo.ai.
MakeAIVideo MCP server
@makeaivideo/mcp
MCP server with 44 tools that lets Claude, ChatGPT, Cursor and any MCP client create finished AI videos and post them to connected social accounts. Hosted at https://mcp.makeaivideo.ai.
MakeAIVideo TypeScript SDK
@makeaivideo/sdk
Official TypeScript and JavaScript SDK for the MakeAIVideo REST API: create AI videos from code and post them to connected social accounts.
MakeAIVideo CLI
@makeaivideo/cli
Command line tool for MakeAIVideo: brief in, finished MP4 out, then post it. Prints JSON for AI agents and CI.
Reference
AI video API reference by group
65 endpoints, grouped by what they act on.
Pricing
AI video API pricing: included on every plan
There is no separate API bill. Renders made through the API draw on the same monthly credit allowance as the app, starting at $9 a month for 450 credits, topped back up on each billing date. New customers start with a 7-day free trial at $0 today. If a month runs heavy, top-up packs add credits that never expire. Compare MakeAIVideo plans and pricing.
AI video API FAQs
Open source packages, authentication, render status and what a render costs.
Something else? Visit the help center
Are the packages open source?
Yes. The SDK, the MCP server and the CLI are MIT licensed and public at github.com/makeaivideo-ai, published on npm as @makeaivideo/sdk, @makeaivideo/mcp and @makeaivideo/cli. Issues and pull requests are welcome on any of them.
Is the API included in my plan, or priced separately?
It is included. API renders draw on the same monthly credit allowance as the app, so there is no separate API bill and no separate rate card. Plans start at $9 a month for 450 credits, topped back up on each billing date.
How do I authenticate?
With a bearer API key created in your account settings. The same key works for the REST API, the CLI and the MCP server. Full detail, including scopes, is on the authentication page.
How do I know when a render has finished?
Subscribe to a webhook and we call you when the render completes or fails. Polling the video endpoint also works, but the webhook is the reason you do not have to.
Can an AI assistant use MakeAIVideo directly?
Yes, through the MCP server. Point an MCP client such as Claude at your account and video generation becomes a tool the assistant can call, with no glue code in between.
Can the API post videos to TikTok, Instagram or YouTube?
Yes. The publishing endpoints, the SDK, the CLI and the MCP server can post a finished video now or on a schedule to connected TikTok, Instagram, YouTube, Facebook Pages, LinkedIn, X, Threads, Pinterest, Bluesky, Telegram and Discord accounts. A person connects each account once in a browser. X takes videos up to 140 seconds.
What does a render cost in credits?
Cost scales with length and the pipeline you choose. A 30-second stock-footage video is about 95 credits and a 30-second video with AI-generated scenes about 428, so the entry plan covers a few short stock-footage videos a month before you would need a top-up pack. The estimate endpoint quotes the exact cost before you render.