Janusly API
Generate images and videos programmatically — the same models, quality, and pricing as the app. Available on all paid plans.
What the API can do
- Generate images — text-to-image and image-to-image with reference images, up to 4 variants per request. Images are plan-included on Team and above and never spend credits.
- Generate videos — image-to-video from 4 to 60 seconds depending on model, with optional audio, charged in the same render credits as the app.
- Upload reference media — images, videos, and audio to steer generations.
- Poll render status — renders complete asynchronously; failed renders refund automatically.
The API does notexpose the chat agent, trend data, or personas — it is a generation surface. Everything you create also appears in the app, in an "API renders" workspace on your account.
Prefer to work from an agent? The MCP server wraps this whole API as tools for Claude Code, Codex, Cursor, and any other MCP client.
Base URL
All endpoints live under https://janusly.com/api/v1 and speak JSON (uploads are multipart form data). Requests authenticate with a bearer API key — see Authentication.
Endpoints
POST /uploads— upload reference images, video, or audioPOST /renders— start an image or video renderGET /renders/{id}— poll a render for completionGET /account— check your plan and render-credit balance
New to the API? The quickstart walks the whole flow in three curl calls.
Versioning
The API is versioned in the path (/api/v1). We add fields and models without notice; breaking changes get a new version. Pin nothing to response-field order, and ignore fields you don't recognize.