create a new pod.

you can create a pod either from a template or by specifying an image directly.

examples:
  # create from template (recommended)
  runpodctl pod create --template-id runpod-torch-v21 --gpu-id "NVIDIA GeForce RTX 4090"

  # create with custom image
  runpodctl pod create --image runpod/pytorch:1.0.3-cu1281-torch291-ubuntu2404 --gpu-id "NVIDIA GeForce RTX 4090"

  # create a cpu pod
  runpodctl pod create --compute-type cpu --image ubuntu:22.04

  # find templates first
  runpodctl template search pytorch
  runpodctl template list --type official

Usage:
  runpodctl pod create [flags]

Flags:
      --cloud-type string          cloud type (SECURE or COMMUNITY) (default "SECURE")
      --compliance string          comma-separated compliance requirements (e.g., HIPAA,SOC_2_TYPE_2)
      --compute-type string        compute type (GPU or CPU) (default "GPU")
      --container-disk-in-gb int   container disk size in gb (default 20)
      --country-code string        limit pod to a specific country (e.g., US, DE)
      --data-center-ids string     comma-separated list of data center ids
      --docker-args string         docker cmd arguments
      --env string                 environment variables as json object
      --global-networking          enable global networking (secure cloud only)
      --gpu-count int              number of gpus (default 1)
      --gpu-id string              gpu id (from 'runpodctl gpu list')
  -h, --help                       help for create
      --image string               docker image name (required if no template)
      --min-cuda-version string    minimum cuda version (e.g., 12.6)
      --name string                pod name
      --network-volume-id string   network volume id to attach
      --ports string               comma-separated list of ports (e.g., '8888/http,22/tcp')
      --public-ip                  require public ip (community cloud only)
      --registry-auth-id string    container registry auth id (from 'runpodctl registry list')
      --ssh                        enable ssh on the pod (default true)
      --stop-after string          auto-stop datetime (e.g., 2026-04-15T00:00:00Z)
      --template-id string         template id (use 'runpodctl template search' to find templates)
      --terminate-after string     auto-terminate datetime (e.g., 2026-04-15T00:00:00Z)
      --volume-in-gb int           volume size in gb
      --volume-mount-path string   volume mount path (default "/workspace")

Global Flags:
  -o, --output string   output format (json, yaml) (default "json")
exit=0
