Skip to main content

​
What is an instance?

A GPU container running your code. Comes with GPU, CPU, RAM, and persistent storage at /workspace.

​
Lifecycle

cassian up   -> running
cassian down -> stopped (workspace saved)

​
Resources

CPU and RAM are assigned automatically based on your GPU. No configuration needed.

​
Idle auto-stop

Instances with no GPU or CPU activity for 30 minutes are automatically stopped. Your workspace is saved before shutdown, so nothing is lost. This means you will not get charged for forgetting to run cassian down.

​
Switching GPU types

Run cassian switch to pick a different GPU. Your workspace volumes are cloud-backed so they attach to any node. You do not need to re-download your data.