Get API key

Vivideo API · v1

Build video into your product

The Vivideo API turns a prompt into a finished video — the same engine behind app.vivideo.ai. Auto-generate complete AI-avatar videos, or call 32 frontier models directly. One REST API, JSON in and out.

32 video modelsAuto & manual modesAvatars · voices · brand kitsSigned webhooksAgent-ready
Get an API keyQuickstart →API reference

Make your first call

Every request is authenticated with a bearer key and returns JSON. Here's your account:

GET /v1/me
curl https://api.vivideo.ai/v1/me \
  -H "Authorization: Bearer $VIVIDEO_API_KEY"

What you can build

Create videos

Auto-generate avatar videos, or call any model directly. Async with polling or webhooks.

32 models

Veo, Sora, Kling, Seedance and more — each with its own options and pricing.

Media & assets

Avatars, voices, brand kits, and file uploads for image-to-video and references.

Webhooks

Signed video.completed / video.failed events with retries.

Credits & billing

Fixed per-model pricing, atomic charges, automatic refunds on failure.

For AI agents

Machine-readable docs, llms.txt, and an OpenAPI spec built for LLMs.

✓
Built for developers and agents alike. Every endpoint returns consistent, typed JSON; every error has a stable code and a next step. The whole API is described in an OpenAPI spec and summarized in llms-api.txt so AI agents can call it reliably. See For AI agents.

How it works

Video generation is asynchronous. You create a video and get an id back immediately; the render finishes in the background (seconds to a few minutes). Track it two ways:

  • Poll GET /v1/videos/{id} until status is completed.
  • Webhooks — register an endpoint and receive a signed event on completion.

Credits are charged from your normal Vivideo balance when a render starts, and refunded automatically if it fails. New here? Start with the Quickstart.

Machine-readable: interactive reference · openapi.yaml · llms-api.txt. Base URL https://api.vivideo.ai.