Skip to main content

Available GPUs

TypeVRAM$/hrBest for
H200141GB$3.59Frontier model training
H100 SXM80GB$2.69Large model training
H100 PCIe80GB$1.99Large model training
A100 SXM80GB$1.39General training
A100 PCIe80GB$1.19General training
L40S48GB$0.79Inference, fine-tuning, video gen
L4048GB$0.69Inference, fine-tuning
RTX 6000 Ada48GB$0.74Workstation workloads
A4048GB$0.35Cost-effective large models
L424GB$0.44Efficient inference, fast iteration
A10G24GB$0.38Good all-rounder
RTX 409024GB$0.34Fast iteration
RTX 309024GB$0.22Budget training
Billed per minute. All users start with $10 free credit.

Selecting a GPU

gpu:
  count: 1
  type: l4
cassian init shows live availability so you only pick what is ready.

Switching GPUs

cassian switch
Interactive picker showing live availability and pricing. Your data moves with you. Volumes are cloud-backed and attach to any node.

Auto-stop

Instances automatically stop after 30 minutes of idle time (no GPU or CPU activity). You won’t get charged for forgetting to run cassian down.

Billing

  • Credits are deducted per minute while your instance is running.
  • If your balance reaches $0, the instance is automatically stopped.
  • If a container crashes, billing stops within 60 seconds.
  • Check your balance anytime with cassian status or in the dashboard.