Prerequisites
Before you install ekai, sign up for a trial and confirm the tools you need for your chosen path.
Sign up and licensing
Open the ekai licensing portal and start your 14-day free trial:
The trial starts as soon as your organization signs up. Anyone joining with the same company email shares that trial. The first person to sign up and install becomes the owner.
To extend, update, or convert to a paid plan, contact sales@ekai.ai or use Talk to Sales in the portal.
Start the install
Once you have signed up, go to licensing.ekai.ai and follow the install instructions shown there.
You will be guided to:
- Grant browser access so your trial can be activated
- Sign in to the licensing portal
- Choose local or GCP
- Install any missing tools (when you approve)
Required CLI tools
For local deployment
Docker is the only requirement for local deployment:
- Docker Desktop 20.10+ (macOS / Windows)
- Docker Engine with Docker Compose (Linux)
Install from the official Docker documentation:
- Docker Desktop (macOS / Windows)
- Docker Engine (Linux)
Windows users must use Windows Subsystem for Linux 2 (WSL2). While installing, users must ensure Ubuntu installation as well and then use the WSL shell to execute the installation script.
For GCP deployment
GCP requires several CLI tools. Missing tools can be installed for you during setup when you approve. To install them yourself, use the official documentation:
| Tool | Official documentation |
|---|---|
| Terraform | Install Terraform |
| Google Cloud CLI (gcloud) | Install the gcloud CLI |
| gke-gcloud-auth-plugin | Install the GKE auth plugin |
| kubectl | Install kubectl |
| jq | Download jq |
| dig | Included with bind-tools / dnsutils on most systems, or via your OS package manager |
GCP project and permissions
For GCP you will need:
-
A GCP project with billing enabled
-
These roles:
- Compute Admin
- Kubernetes Engine Admin
- Service Account Admin
- Cloud SQL Admin
- DNS Administrator
-
An authenticated gcloud CLI:
gcloud auth login
- A domain (or subdomain) to configure the deployment and host ekai’s client-facing web app.
System requirements
Minimum
- CPU: 4 cores
- RAM: 8 GB
- Disk: 20 GB available
Recommended
- CPU: 8 cores
- RAM: 16 GB
- Disk: 50 GB available
Next steps
- Local deployment — Run ekai with Docker
- GCP deployment — Run ekai in your GCP project



