API · Video
AI Video Generation API for Production Workflows
Run text-to-video and image-to-video pipelines through one observable API with idempotency, progress tracking, and CDN outputs.
Built for agents and applications
Text-to-video and image-to-video workflows
Multiple video models without rewriting your integration
Asynchronous runs with polling and progress events
Final MP4 output on imageat's CDN
const response = await fetch(
"https://api.imageat.com/v1/workflows/wf_video_production/runs",
{
method: "POST",
headers: {
Authorization: `Bearer ${process.env.IMAGEAT_API_KEY}`,
"Content-Type": "application/json",
"Idempotency-Key": crypto.randomUUID()
},
body: JSON.stringify({
// Uses the nodes and edges saved in the published workflow.
nodes: [],
edges: []
})
}
)
const run = await response.json()
// Poll GET /v1/runs/{run_id} or subscribe with SSE.
console.log(run.run_id)How it works
From instruction to production asset.
Build and publish a workflow
Choose the model, inputs, edit steps, and output once in imageat's visual workflow builder.
Create an API key
Generate a server-side iat_live_ key and keep it outside browser code and source control.
Run, observe, and receive outputs
Start runs with idempotency, follow progress by polling or SSE, and store the returned CDN URLs.
Frequently asked questions
Is this a direct model API?
imageat provides one API layer over text-to-video and image-to-video workflows. You can change the model or add processing steps without rebuilding the calling application.
How is usage billed?
Runs use the credits on the imageat account connected to the API key or MCP session. The required credits depend on the selected model and workflow.
Can I use outputs commercially?
Commercial use depends on imageat's terms and the terms of the models and source assets used in your workflow.
Build once. Call it from anywhere.
Start with the hosted interface, then move the same visual workflow into your agent or production application.
Create an API key