Core concepts
Inference-time insertion
Inference-time insertion places sponsorship near the user experience while the application remains in control of timing and rendering.
Technical model
The app creates a job, receives slot IDs, polls decisions, and renders assets when the placement is ready. The model response remains owned by the publisher app.
Rendering
Use the Script Tag for automatic rendering or the Server API for custom rendering. Either path records beacons for proof and settlement.
Fallback
If no decision is ready, keep the user flow moving. Do not block inference completion on a sponsorship response unless your product explicitly wants that behavior.
Need rollout review?
Start in the dashboard, choose Script Tag or Server API, and use contact only when you need rollout review, enterprise coordination, or non-standard integration help. Billing beacon rules live in the API concepts guide.