See your GPU waste in one command.
piqc is our open-source scanner, not the platform. Point it at a cluster and get an instant cost report in your terminal: a glimpse of what Paralleliq sees continuously, not a trial of the product itself.
Run it
Kubernetes Job, pip/pipx, or Docker: pick whichever fits your workflow. Read-only; it makes no changes to your cluster.
Read the report
A cost table per deployment, plus a summary: unallocated nodes, tier misplacement, idle spend, right in your terminal, no dashboard needed.
See what's next
This is a one-time snapshot. The platform adds continuous monitoring, dollar-tracked recommendations, and approval-gated fixes across your whole fleet.
Install piqc, three ways
Runs inside your cluster as a one-off Job, using its own scoped RBAC role. The third command shows the report. Self-deletes after 10 minutes.
kubectl apply -f https://raw.githubusercontent.com/paralleliq/piqc/main/deploy/rbac.yaml kubectl apply -f https://raw.githubusercontent.com/paralleliq/piqc/main/deploy/scan-job.yaml kubectl logs -f job/piqc-scan -n kube-system
Reads your existing kubeconfig. Use pipx if you have any pydantic v1 dependencies elsewhere; it avoids the conflict.
pipx install piqc piqc scan
No local Python needed: mount an exported static kubeconfig into the container.
docker run --rm -v ~/.kube/config:/root/.kube/config:ro ghcr.io/paralleliq/piqc scan
All three produce the same read-only scan and a cost report in your terminal, with no account needed. Full options and flags are in the piqc README.
What's included
- Free and open source, forever
- No account, no credit card
- Read-only, makes no changes
- One-time snapshot, not continuous monitoring