Hyperstack On-Demand Kubernetes
Kubernetes clusters on GPU-enabled worker nodes with NVIDIA-optimised images, created and scaled through the console or the API, for training jobs, fine-tuning pipelines and inference services alongside general cloud-native applications; worker nodes are billed hourly.
Find alternatives to Hyperstack On-Demand Kubernetes
What it does
- GPU cloud
- Model training
- Model inference
Sources
- Pricing
-
Product page, How Billing Works: "Hyperstack bills Kubernetes clusters based on the compute resources powering your worker nodes." and "Worker Nodes : Charged hourly based on the selected flavour."; no rate is printed on the page, which sends readers to the billing documentation.
- Description
-
Official documentation · 16 Nov 2026
Product page: "Hyperstack On-Demand Kubernetes provides a fast, flexible environment for deploying, scaling and managing production-ready Kubernetes clusters built for AI and cloud-native applications."
- What it does
-
Official documentation · 16 Nov 2026
Product page: "Hyperstack On-Demand Kubernetes provides a fast, flexible environment for deploying, scaling and managing production-ready Kubernetes clusters built for AI and cloud-native applications."
- Sold within
-
Product page: "Master, Bastion and Load Balancer Nodes : Included at no additional resource cost." The page names no plan it is included in or bought on top of, so access is not-found.
- Name
-
Official documentation · 16 Nov 2026
Product page: "Hyperstack On-Demand Kubernetes provides a fast, flexible environment for deploying, scaling and managing production-ready Kubernetes clusters built for AI and cloud-native applications."
- URL
-
Official documentation · 16 Nov 2026
Product page: "Hyperstack On-Demand Kubernetes provides a fast, flexible environment for deploying, scaling and managing production-ready Kubernetes clusters built for AI and cloud-native applications."
- Deployment
-
Official documentation · 16 Nov 2026
Product page: "Hyperstack On-Demand Kubernetes provides a fast, flexible environment for deploying, scaling and managing production-ready Kubernetes clusters built for AI and cloud-native applications."
- Interfaces
-
Official documentation · 16 Nov 2026
Product page: "They can be created, scaled or deleted through the Console or API."
Also from Hyperstack
-
Hyperstack Infrastructure service
On-demand GPU cloud billed per minute, with NVIDIA H100, H200, B200 and B300 virtual machines, reserved pricing, spot VMs, private clusters, object storage and an AI Studio, driven from a web console, a Python and a Go SDK and a Terraform provider; per-hour prices for each GPU are published.
-
Hyperstack AI Studio Platform
Gen AI platform on Hyperstack's GPU cloud for fine-tuning open-source LLMs from uploaded data, evaluating them, testing them in a playground and serving them through serverless or dedicated inference; billed per token and per fine-tuning minute.
Open Hyperstack in the directory
Something wrong here? Send a correction — quote this product id: hyperstack-on-demand-kubernetes.