Resources
Guides
Step-by-step guides for common jobs with the Tovano API. Each one uses real requests you can run with your own key.
Generate images
Create images from a text prompt, choose size and count, wait for the result and read the charge.
image/demo-1 · POST /v1/tasks
Generate video and speech
Make short video clips charged per second and speech charged per 1,000 characters, and work out the cost before you send.
video/demo-1 · audio/demo-tts-1
Stream chat replies
Call text models with the chat completions format, stream the reply token by token and read the exact charge from the last chunk.
text/demo-chat-1 · POST /v1/chat/completions
Run an agent job
Send one brief to agent/ad-kit-1 and get a headline, body copy and a matching image, with each step and its charge listed.
agent/ad-kit-1 · POST /v1/tasks
Receive results with webhooks
Stop polling. Get every finished task pushed to your server, check the signature and handle retries and duplicates safely.
webhook_url · Tovano-Signature
Track spend per end customer
Tag every call with your own customer ID, then read spend per customer from the API for re-billing, margin checks and fair-use limits.
X-Tovano-Customer · GET /v1/usage?group_by=customer
Set spend limits and low-balance alerts
Cap monthly spend, get warned before the balance runs out, keep production running with auto top-up, and handle the 402 errors in code.
Billing settings · 402 insufficient_credits · 402 spend_limit_reached
Handle errors and retry safely
Read the error format, retry only what is safe to retry, respect rate limits, and use idempotency keys so a retry never creates a second charge.
Idempotency-Key · Retry-After · X-Request-Id
Looking for the full reference?
Every endpoint, field and error code is in the documentation.