01
Send the request
POST your generation request — text prompt, image input, and target model family — to a single API endpoint. No per-model SDK integration needed.
NewCinema Studio is live · 61 models across 25 families in one engine · Start free — no credit card
Production — ORVIK
61 models. One API.
Orvik routes your generation requests across 61 frontier image and video models in parallel — Kling, Veo, Wan, FLUX, and more — with a single unified API and webhook delivery.
STORYBOARD — HOW IT WORKS
Three steps. Done.
- 02
Engine dispatches
The engine routes to the optimal model, handles retries and quota management, and runs parallel jobs when you request multiple variants — all invisible to your application.
- 03
Receive via webhook
When the job completes, the engine sends a webhook to your endpoint with the output URL and metadata. Poll the run ID or stream updates via SSE — your choice.
SCREENING ROOM — SAMPLE OUTPUT
See what the engine produces.
Sample output videos and images — representative of format categories available in Orvik. Hover or scroll to play video tiles.
Sample Video16/9
Sample Image16/9
Sample Image16/9
Sample Image9/16
Model Router1/1
Parallel Dispatch1/1
CALL SHEET — WHAT YOU GET
Built to perform.
| No. | Item | Detail |
|---|---|---|
| 01 | 61 frontier models, one integration | Kling, Veo, Wan, FLUX, Sora, MiniMax, Pixverse, LTX, and more — all accessible through a single API. The catalog updates as new models ship; your integration never changes. |
| 02 | Parallel dispatch for multi-variant jobs | Generate 6 ad variants in the same time it takes to generate one. The engine fans out parallel generation requests across the model fleet and collates results. |
| 03 | Webhook and SSE delivery | Receive completions via webhook POST to your endpoint, or stream status updates with SSE. No polling loops, no lost results — the engine tracks every job. |
| 04 | Retry, quota, and cost management built-in | The engine handles model quota limits, automatic retries on transient failures, and cost-gate controls so expensive jobs never run without approval. |
Real output — straight off the engine
22 real renders. No mockups, no stock — every tile was generated here.
Powered by
The rate card
Start free. Scale when it works.
Creator
$9 per month
Studio
$29 per month
Studio Max
$79 per month
Studio
$29 per month
Compare all plans
Creator
$9 per month
Studio Max
$79 per month
FAQ
Questions, answered.
A unified REST API that routes image and video generation requests to 61 frontier models. You POST a job, the engine dispatches it to the right model, and delivers the result via webhook or SSE — without per-model SDK integrations.
Kling 3.0, Veo 3.1, Wan 2.7, FLUX Dev / Kontext, MiniMax, Pixverse, LTX, and more — 61 models across 25 families. The catalog is updated as new frontier models release — your integration remains unchanged.
Two options: (1) register a webhook URL and receive a POST when the job completes; (2) poll the job status endpoint or open an SSE stream for real-time updates. Both return the output asset URL and metadata.
Yes — set the count parameter to generate N variants of the same job in parallel. The engine dispatches all variants simultaneously and delivers each result individually via webhook as they complete.
Yes. The engine enforces per-job and per-period cost gates. Jobs that would exceed the configured budget limit are queued or rejected before any generation credit is spent.
API access is included in Studio and Enterprise plans. Documentation and an API key are available from the account settings page after sign-up.
Yes — plans start at $9/month for Creator and $29/month for Studio. Enterprise pricing is available on request.
Keep exploring
LAST LOOKS
Every frame, one engine.
A closing reel of real engine output — video and image, across formats and use cases. Every tile is a sample render.