Skip to main content
Modify the resources of a running instance without creating a new one. This allows you to scale up or down based on your workload needs.

Modify an Instance

Use the modify command with the instance ID and new configuration:
tnr modify <instance_id> [options]

Available Options

OptionDescription
--gpuGPU type (e.g., a6000, a100, h100)
--vcpusNumber of virtual CPUs (prototyping mode)
--disk-size-gbDisk size (can only be increased)
--modeSwitch between prototyping and production

Examples

# Switch to a different GPU
tnr modify 0 --gpu a100

# Scale up CPUs (RAM scales automatically at 8GB per vCPU)
tnr modify 0 --vcpus 16

# Increase disk size
tnr modify 0 --disk-size-gb 300

# Switch to production mode
tnr modify 0 --mode production
Disk size can only be increased, not decreased.

What Can Be Modified

ResourceCan Modify?Notes
GPU TypeYes
GPU CountYes
vCPUsYesPrototyping mode only
RAMYesScales with vCPUs (prototyping) or GPUs (production)
ModeYesSwitch between prototyping and production
Disk SizeIncrease onlyCannot shrink disk
RAM is automatically determined based on your configuration:
  • Prototyping mode: 8GB per vCPU
  • Production mode: 90GB per GPU