Skip to main content
Deployment is not yet available. We are onboarding partners on a case-by-case basis. Contact us to get access.
Once your model runs locally with reactor run, you’re ready to deploy. Reactor handles GPU provisioning, container orchestration, and global routing. Your model is live and accepting sessions in under 3 minutes.

Deployment flow

  1. Authenticate - Log in with your API key and configure registry access.
  2. Register your model - Provision storage and a container registry for your model.
  3. Upload weights - Transfer weights from S3 or your local machine.
  4. Push your Docker image - Build and push a container with your model code.
  5. Publish - Deploy to Reactor’s GPU cloud.

Prerequisites

  • A working model that runs locally with reactor run (see Development)
  • A Reactor partner account and API key (request access at reactor.inc)
  • Docker installed and running

Next

Authentication

Log in and configure registry access.

Model Management

Register your model and manage versions.