Built for developers

One POST request.
One MP4.
No FFmpeg.

JSON in, MP4 out. Official SDKs for Node.js, Python, PHP, Ruby, Go, and .NET — or plain HTTP and curl.

MCP server for AI agents — Claude, ChatGPT, any MCP client. MCP docs
npm install @ilovevideoeditor/sdk-node
1import { ILoveVideoEditorClient } from '@ilovevideoeditor/sdk-node';
2
3const client = new ILoveVideoEditorClient({ apiKey: 'vf_live_xxx' });
4
5const result = await client.render(
6 { name: 'Hello', layers: [...] },
7 {
8 onProgress: (s) => console.log(`${s.status} — ${s.progress}%`),
9 },
10);
11
12console.log(result.downloadUrl);

Self-hosted FFmpeg vs iLoveVideoEditor API

Spoiler: one requires a DevOps team, the other requires 5 minutes

FFmpeg Pipeline
iLoveVideoEditor API
Setup time
2 days
5 minutes
Lines of code
~500
~20
Render speed
3x realtime
1x realtime
Monthly cost
$200+ server
Pay per render
Maintenance
High
None

Estimates based on a typical self-hosted FFmpeg setup.

Browser or Cloud

Two ways to render, same JSON format

Browser

Client-side rendering with WebCodecs. Instant preview, zero server cost, no upload. Perfect for in-app editors and tools.

Free
Real-time preview

Cloud

Server and edge rendering via the REST API. Highest quality, webhook support, and auto-scaling. Pay per render.

Best quality
Auto-scaling
Recommended for backends

Render any template via API

All 259 motion templates are renderable programmatically. Pick a template, pass its variables as JSON, and get back an MP4.

bash
1curl -X POST https://api.ilovevideoeditor.com/v1/templates/saas-launch/render \
2 -H "Authorization: Bearer YOUR_API_KEY" \
3 -H "Content-Type: application/json" \
4 -d '{
5 "variables": {
6 "headline": "Ship 10x faster",
7 "subhead": "Rendered from a template"
8 }
9 }'

Or use plain HTTP

Don't use our SDK? No problem. Here's a complete request in curl.

bash
1curl -X POST https://api.ilovevideoeditor.com/v1/render \
2 -H "Authorization: Bearer YOUR_API_KEY" \
3 -H "Content-Type: application/json" \
4 -d '{
5 "videoJson": {
6 "duration": 3,
7 "width": 1920,
8 "height": 1080,
9 "layers": [
10 {
11 "type": "solid",
12 "color": "#e5352c",
13 "duration": 3
14 }
15 ]
16 }
17 }'
6
languages supported
20
lines of code to first render
0
infrastructure to manage

Frequently asked questions

Which languages have official SDKs?

Node.js, Python, PHP, Ruby, Go, and .NET, plus a plain REST API and a public OpenAPI spec for everything else.

How is rendering billed?

Pay per render — no servers to provision, no infrastructure to maintain, and no monthly commitment required to start.

Can AI agents trigger renders?

Yes. The MCP server exposes the same video tools to AI agents, and ready-made integrations exist for Zapier, Make, and n8n.

How do I authenticate API calls?

With a Bearer API key from the dashboard. The docs cover authentication, rate limits, and render webhooks.

Ship video features this week

Your PM will think you're a wizard. Your infrastructure team will thank you.