Overview
A blazing-fast, future-proof command line for provisioning and orchestrating everything on Huddle Cloud.
hudl is the official command-line interface for Huddle Cloud. One static binary, zero runtime dependencies, every endpoint of the platform mapped to a single predictable grammar.
It's the same control plane that powers the dashboard, but compressed into something you can pipe, script, alias, and live in.
1:1 with the API reference
Every command in this section maps directly to a documented endpoint in the Cloud API reference. If you can read the API docs, you already know hudl.
The grammar
Every hudl command follows the same shape:
hudl <resource> <verb> [target] [flags]If you know one resource, you know them all.
Cloud resources
| Resource | Maps to | Example |
|---|---|---|
vm | Instances | hudl vm create web-01 |
volume | Volumes | hudl volume attach vol-01 --to vm-01 |
fip | Floating IPs | hudl fip associate 203.0.113.10 --to vm-01 |
sg | Security Groups | hudl sg create web --description "HTTP/HTTPS traffic" |
network | Networks | hudl network create vpc-prod --cidr 10.0.0.0/16 |
key | Key Pairs | hudl key create laptop --public-key ~/.ssh/id_ed25519.pub |
flavor | Flavors | hudl flavor list |
image | Images | hudl image list |
region | Regions | hudl region list |
GPU resources
| Resource | Maps to | Example |
|---|---|---|
gpu deploy | GPU Deployments | hudl gpu deploy --offer ofr_8xH100 |
gpu offers | GPU Offers | hudl gpu offers |
gpu key | GPU SSH Keys | hudl gpu key upload prod-key --file ~/.ssh/id_ed25519.pub |
gpu volume | GPU Volumes | hudl gpu volume create datasets --size 500 --type nvme |
gpu webhook | GPU Webhooks | hudl gpu webhook create https://example.com/hook |
gpu apikey | GPU API Keys | hudl gpu apikey create ci-bot |
gpu waitlist | GPU Waitlist | hudl gpu waitlist add |
Global flags
These flags work with every command:
| Flag | Short | Description |
|---|---|---|
--region <code> | Override the active region for this call. | |
--workspace <slug> | Override the active workspace for this call. | |
--output <format> | -o | Output format: table, json, yaml, wide, name. |
--idempotency-key <key> | Attach an idempotency key for safe retries (mutating commands only). |
Quickstart
# 1. install
curl -fsSL https://get.huddle01.com/hudl | sh
# 2. authenticate with your API keys
hudl login --token mk_xxx --gpu-token gk_xxx
# 3. set workspace and region context
hudl ctx use acme-prod
hudl ctx region eu2
# 4. provision a VM
hudl vm create web-01 \
--image ubuntu-24.04 \
--flavor c2.small \
--boot-disk-size 50 \
--key laptop \
--sg default \
--public-ipEnd-to-end example
A common workflow: create a VM, attach storage, assign a public IP, and open port 443.
# create a security group and open HTTPS
hudl sg create web-tier --description "Public web servers"
hudl sg rule add <sg-id> --protocol tcp --from-port 443 --to-port 443 --cidr 0.0.0.0/0
# launch the VM
hudl vm create api-server \
--image ubuntu-24.04 --flavor c4.large \
--boot-disk-size 40 --key deploy-key \
--sg web-tier --public-ip
# create and attach a data volume
hudl volume create api-data --size 100 --type ssd
hudl volume attach <volume-id> --to <vm-id>
# assign a floating IP
hudl fip associate 203.0.113.42 --to <vm-id>Scripting and automation
hudl is built for scripting. Use --output name to get just IDs, --output json to pipe into jq, and idempotency keys for safe retries in CI pipelines.
# list all VM IDs in a region
hudl vm list --region us1 -o name
# get a VM's status with jq
hudl vm get <id> -o json | jq -r '.status'
# safe retry in CI
hudl vm create staging-01 \
--image ubuntu-24.04 --flavor c2.small \
--boot-disk-size 30 --key ci-key --sg default \
--idempotency-key "deploy-$(git rev-parse --short HEAD)"Where to next
Install
Get the binary on macOS, Linux, Windows, or Docker.
Authentication
API keys, workspace contexts, and CI setup.
Configuration
hudl.toml, env vars, output formats, shell completions.
Virtual Machines
The most common starting point — provision and manage VMs.
Volumes
Create, attach, and manage block storage.
Security Groups
Firewall rules for your instances.
GPU Deployments
Deploy and manage GPU instances.