TL;DR — Run a ready-made image or video style with one HTTPS call from your server, then poll for the result or receive a signed webhook. You never pick a model; you only pay for a run that delivered. Test keys (
wdr_test_…) return a sample for $0. Base URL:https://api.wandura.ai.
This is the API for generating images and videos with Wandura styles. (Streaming licensed films to an app is a different API: see the Delivery API.)
Pick your path
Two ways to call the same engine
Start with jobs. New integrations use the Jobs API: one door, one set of words for statuses and errors. Runs stay supported and run on the same engine; nothing you built on them changes.
| Jobs (recommended) | Runs | |
|---|---|---|
| Submit | POST /v2/jobs |
POST /v2/styles/{style} |
| Result | A job: job_id, GET /v2/jobs/{job_id}, signed webhook_url |
A run: status_url, cancel_url |
| Statuses | queued, processing, succeeded, failed, cancelled, timeout |
The same words in state (status keeps the older words) |
| Best for | Every new integration: apps, backends, workflow engines | Integrations already built on runs |
| Same underneath | Keys, wallet, prices, fallback, and no charge for a run that did not deliver |
Rolling out: the Relax lane (lane) and a style's own inputs (inputs) on POST /v2/jobs, so every style a key can run through runs also runs as a job (see Jobs API).
Reference
- OpenAPI 3.1 for the whole API: api.wandura.ai/v2/openapi.json (no key needed).
- One style's inputs:
GET /v2/public/styles/{style}/openapi.json. - MCP server for agents that act at run time:
https://api.wandura.ai/mcp(same Bearer key). - Status and maintenance: wandura.instatus.com.
Updated 16 Oct 2026 · Change history (2)