For developers

One API for every model. Agent-ready.

Call any model in the catalog with a single key and one credit balance — or plug Vidova into Claude, Cursor and other AI agents with the built-in MCP server.

Read the docs Start free

No credit card required

What you can do

One REST API
List models, generate, poll and download results with the same request shape for video, image, audio and avatars.
MCP server
Agents can list models, estimate the cost, generate media, check your balance and start Autopilot videos.
Know the price first
estimate_cost and get_price return exact credits before anything runs — no surprises in automated pipelines.
Same credits, same library
API generations use your plan's credits and appear in your library, ready for the editor.

How it works

1
Create an API key
On the Studio plan, open Developers in the app and create a key.
2
Connect
Call the REST endpoints, or add the MCP server to Claude, Cursor or any MCP client with one config block.
3
Automate
Generate at scale, wire it into your product, or let an agent produce videos end-to-end.

Models behind it

Prices are in credits at default settings — the studio shows the exact price before you generate.

See all models

Questions & answers

Which plan do I need?

API and MCP access are included in the Studio plan. Requests use the same credits and prices as the studio, so you can prototype in the app and automate with the API.

Which MCP clients are supported?

Any client that speaks Streamable HTTP — including Claude Desktop, Claude Code and Cursor. Setup snippets are in the docs.

Are there rate limits?

Yes, sensible per-key limits protect the service; the docs list them along with error codes and retry guidance.

Webhooks or polling?

Both. Pass a webhook_url to be notified when a generation finishes, or poll it by id; results include a URL for every output.