Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.expanse.sh/llms.txt

Use this file to discover all available pages before exploring further.

​
1. Install the CLI

curl -fsSL https://expanse.org.uk/install | sh
Verify: 
expanse version

​
2. Sign in

expanse login
The CLI walks you through the SSO flow and stores a personal API key locally. Confirm it worked: 
expanse status

​
3. Register a compute

A compute is anywhere expanse-daemon runs: a SLURM cluster, Nomad cluster, Kubernetes namespace, Databricks workspace, YARN cluster, cloud batch project, SkyPilot VM, or standalone cloud box. 
expanse compute register
The command asks what kind of compute you’re registering and prints the install command or bootstrap snippet to run on the target. Within a minute the compute appears at console.expanse.sh.

​
4. Run a workload

For scheduler-backed installs, every workload that runs on the compute is captured automatically. No change to how anyone submits.
For environments without a scheduler, wrap the command you want tracked:
expanse run -- python train.py
This is the direct-run path. Scheduler-backed computes capture workloads automatically.

​
5. Predict and diagnose

Right-size resources before you submit: 
expanse analyse train.slurm     # SLURM batch script
expanse analyse train.py        # source file
Get solution-oriented guidance after a failure: 
expanse diagnose <execution-id>
<execution-id> is the Expanse execution ID, or the SLURM job ID, or the Kubernetes pod UID. Whichever you have to hand.

​
Next steps

Architecture

How Expanse captures evidence and trains its own models.

Computes

What a compute is and what’s supported.

Intelligence

How analyse and diagnose work.

CLI reference

Every command and flag.

Authentication

Personal keys and SSO.