Portable OCI images
Pin any compatible public or private-registry image reference and its start command.
Deployment Templates
Configure once. Deploy the same way every time.
A template is a private, reusable workload definition: Docker image, start command, container disk, local volume, mount path, exposed ports and non-secret environment variables.
Transparent baseline
No extra fee
template records are workspace-scoped
Pod
GPU and cluster mode
Serverless
Worker mode
3
NVIDIA · AMD · CPU
EU regions, encrypted storage and network isolation options
Why GPURento
Less infrastructure work. More model progress.
Runtime configuration
Save disk sizes, mount path, HTTP/TCP ports and non-sensitive environment settings.
Repeatable deployments
Reuse one Pod template when creating clusters so every node starts from the same image.
From code to capacity
One clean path to production.
1
Choose workload type
Create a GPU Pod / Cluster template or a Serverless worker template.
2
Save the runtime
Enter image, command, disks, ports and non-secret environment variables.
3
Reuse and update
Select the template during cluster creation, edit it later or delete it without removing existing resources.
deploy.py
await fetch("/api/templates", {
method: "POST",
headers: { "content-type": "application/json" },
body: JSON.stringify({
name: "pytorch-workspace",
workloadType: "pod",
category: "NVIDIA",
image: "pytorch/pytorch:2.6.0-cuda12.6-cudnn9-runtime",
startCommand: "sleep infinity",
containerDiskGb: 50,
volumeGb: 100,
volumeMountPath: "/workspace",
ports: "8888/http\n22/tcp",
environmentVariables: "NCCL_DEBUG=INFO"
})
})✓ Configuration saved to the workspace · provisioning state visible
Included
A serious platform, not a thin wrapper.
Every workload gets the same API, identity, billing and observability layer.
Private workspace ownership
Pod and Serverless types
NVIDIA, AMD and CPU categories
Docker image and start command
Container and local volume disks
Linux mount path
HTTP and TCP ports
Non-secret environment variables
Can users create templates without contacting sales?+
Configure resources directly in the dashboard. Prepaid usage uses wallet credit; monthly rentals have a separate checkout.
Do templates allocate a GPU?+
No. A template stores a reusable configuration. Compute is allocated only when a resource is created from it.
Should API keys be stored in a template?+
No. The environment field is for non-sensitive values only. Passwords, private keys and API tokens need a dedicated secret store.
Can a template be used by a cluster?+
Yes. Pod templates appear in the cluster creation form and provide the image used by every node.
Ready when you are