Cinematic AI video,
one API call away
Build the same engine that powers Omnira Studio into your own product. Text, images, and footage in — production-ready renders out, with webhooks, SDKs, and credits you can reason about.
curl https://api.omnira.art/v1/video/generate \
-H "Authorization: Bearer $OMNIRA_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "omnira-video-3.0",
"modality": "t2v",
"prompt": "Cinematic drone shot over a neon city at night, rain-slick streets",
"resolution": "1080p",
"duration": 8,
"webhook_url": "https://yourapp.com/hooks/omnira"
}'
# → { "jobId": "job_a1b2c3", "status": "queued", "estimatedCredits": 180 }Everything the studio can do — headless
The API exposes the full Omnira pipeline so you can embed generation anywhere.
One endpoint, every model
A single generate call routes across Wan, CogVideoX, and commercial frontier models. We pick the right GPU and model for the job.
Realtime job events
Stream progress over webhooks or SSE — queue.position, job.progress, job.completed, job.failed — and react the moment a render lands.
Predictable credits
Every request returns its exact credit cost up front. Meter usage per key, set spend caps, and reconcile against a double-entry ledger.
Built for scale
Rate limits that grow with your tier, multi-region GPU routing, and signed CDN URLs for every output. From prototype to production.
Core endpoints
REST over HTTPS. Authenticate with a Bearer API key.
| Method | Endpoint | Description |
|---|---|---|
| POST | /v1/video/generate | Queue a text→video, image→video, or video→video render. |
| GET | /v1/video/jobs/:id | Poll a job for status, progress, and the signed output URL. |
| GET | /v1/video/jobs | List and filter your generation history with cursor pagination. |
| POST | /v1/images/generate | Render single images or connected image series in 2K / 4K. |
| POST | /v1/webhooks | Register an endpoint to receive realtime job events. |
| GET | /v1/account/credits | Check the remaining credit balance for an API key. |
Rate limits that scale with you
Prototype and test with shared GPU capacity.
Priority queue and higher concurrency for production apps.
Dedicated throughput, private clusters, and a custom SLA.
Need custom volume? Talk to our team →
Ship your first render today
Create a key, install the SDK, and generate your first clip in under five minutes.