The best prices
Pay only for generations, no monthly plan. 720p for 5 seconds starts at $0.39. The margin stays in your product.
One predictable API for video generation: current models, a price before you start, and async jobs. Wan 3 is in the catalog now — new models join the same integration.
curl -X POST \
https://api-platform.gensta.ai/v1/generations \
-H 'Authorization: Bearer $GENSTA_API_KEY' \
-H 'Content-Type: application/json' \
-d '{
"model": "wan-3.0",
"mode": "text_to_video",
"parameters": {
"resolution": "720p",
"duration_seconds": 5
},
"input": {
"prompt": "A product shot in a sunlit studio"
}
}'{
"id": "a931…",
"status": "queued",
"price": { "amount": "0.39", "currency": "USD" }
}We handle models, the queue, and pricing — you call one contract.
Pay only for generations, no monthly plan. 720p for 5 seconds starts at $0.39. The margin stays in your product.
We start with Wan 3. The catalog will grow, and your code stays the same POST /v1/generations.
Jobs do not get lost in the queue. The price is fixed before launch. A retry does not charge twice.
Text, frames, references, and audio — one contract. You do not rewrite the integration when the next model arrives.
No minimum package. The example below is text-to-video, 720p, 5 seconds.
720p · 5 seconds · text-to-video
Start in the portalDetails live in the docs. This is what you need before the first request.
DocumentationThe API currently includes Wan 3: text, image, references, edit, and extend. New models join the same contract without changing your integration.
Open the portal, create a project and key, then send POST /v1/generations. Read status and the video URL with a separate GET.
Pay as you go from a USD balance. Top up in the developer portal — no monthly plan.
No. Video creation is async: POST returns an ID and queued status immediately. You fetch the result with a later request.
Open the portal, get a key, and run the first generation.