$ cat /models/veo

Veo API — AI video generation without a Google Cloud account

Generate video with Google Veo 3.1 through one async endpoint: veo3.1-fast and veo3.1-pro tiers, 720p to 4K output.

Text-to-video, first/last frame control and up to 3 reference images. Billed only for successful renders.

$ cat /pricing

Seedance 2.0 720p	$0.2000 -34%	per second
Seedance 2.0 1080p	$0.5000 -27%	per second

$ cat example.sh

curl -X POST https://sosana.art/api/veo/create-async \
  -H "Authorization: Bearer $SOSANA_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "drone flight over a neon city at night",
    "model": "veo3.1-fast",
    "aspect_ratio": "16:9"
  }'

# poll until status is COMPLETED
curl https://sosana.art/api/veo/{uid} \
  -H "Authorization: Bearer $SOSANA_TOKEN"

$ cat /features

fast & pro tiers veo3.1-fast for drafts and volume, veo3.1-pro for final quality — both up to 4K
frame control set first and last frames or pass reference images to steer the clip
async + webhooks video renders take minutes — submit the task and get a webhook when done
single token same token and balance as image and text models

$ cat /faq

▸ What is the Veo API?

Veo 3.1 is Google's video generation model. Sosana exposes it via an async API — no Google Cloud setup, single prepaid balance.

▸ What can I control?

Prompt, aspect ratio, first and last frame images, and up to 3 reference images per request.

▸ How long does a render take?

Typically a few minutes depending on tier and resolution. Use webhook_url to get notified instead of polling.

▸ How is it billed?

Prepaid balance, charged per successful render only. Failed renders are retried and never double-billed.

$ ls /models

/models/nano-banana · /models/veo · /models/claude

$ cd / · docs · playground