Documentation

Deployments

How BotPhi provisions, scales, and manages your bot instances.


When you deploy a bot, BotPhi provisions an isolated compute instance, connects it to your configured model and channels, and starts serving requests. Each deployment is immutable — to change config, you create a new deployment.

How Deployments Work

  1. Configuration is validated and encrypted secrets are stored
  2. An isolated compute instance is provisioned in the nearest region
  3. Model connection is established and health-checked
  4. Channel webhooks are registered and verified
  5. The bot starts accepting messages

Regions

BotPhi deploys to the region closest to your users for minimal latency. Available regions:

  • ap-south-1 — Mumbai, India
  • us-east-1 — Virginia, USA
  • eu-west-1 — Ireland, EU
  • ap-southeast-1 — Singapore

Scaling

BotPhi auto-scales based on message volume. You don't need to configure instances or capacity — scaling is handled transparently. Each bot gets dedicated resources, so noisy neighbors don't affect your performance.

During early access, all bots are deployed with standard compute. Custom instance sizes will be available in the Pro plan.