
Images and video, rendered as written
Generate professional-quality content with image and video models, through the API or straight from a model's own page.
Models
Image and video models
How it works
Three steps from key to output
Get an API key
Add credits in the console and create a key. Every request carries it as a bearer token, and the key prefix tells you it is ours.
Start a generation
Post a model id, your prompt and the size or resolution you want. The call returns a record id right away; the model runs asynchronously.
Download the output
Read that record back until it is finished, then download the file. If a generation fails the credits are refunded automatically.
curl https://vellria.com/v1/generate/image \
-H "Authorization: Bearer vll_..." \
-H "Content-Type: application/json" \
-d '{
"model": "vellria-lite",
"prompt": "A cat asleep in space, cinematic light",
"size": "1024*1024",
"seed": 42
}'Why Vellria
What you get
FAQ
Frequently asked questions
How do credits work?
1 credit = $0.01. Credits are deducted when a generation starts and refunded automatically if it fails. Each model's page shows what it costs.
How long are my generations kept?
Files are kept for 14 days and then deleted from the server. Your billing record stays but the file goes — download anything you want to keep.
Is there an API?
Yes. Get your key from the console and call every model with the same body; the API reference has the endpoints and a curl example. Generation starts asynchronously and the result is read back from the same record.
How do payments work?
No subscription. You top up with a card or with crypto — on the crypto side USDT · TRC20 is recommended — and credits are added automatically once the payment confirms. Each method has its own packages, so the same amount buys different credits.






