TABLE OF CONTENTS
NVIDIA H100 SXM On-Demand
We’re back with your monthly dose of Hyperstack updates!
September has been a busy month at Hyperstack, packed with updates to make your AI workflows smoother. This month also brings key Kubernetes enhancements, giving you even more control and efficiency in managing your clusters.
New on Hyperstack Platform
Check out what's new on the Hyperstack cloud platform this month:
Change Password via UI
You can now easily change your account password from the Security page in Hyperstack, ensuring faster and more secure password management. Learn more.
Virtual Machine Console Logs
You can now access VM console logs directly through the UI or API. Whether you’re troubleshooting or monitoring performance, this makes it faster and easier to get the insights you need without extra steps.
Move Volumes Across Environments
Need to transfer volumes within the same region? Now you can move them seamlessly via both UI and API.
Volume Attachment Protection
Protect your workloads from accidental disruptions. New API options prevent volumes from being detached from VMs, whether after volume attachment or during volume attachment.
New in Kubernetes
We made major enhancements to make your Kubernetes experience more powerful, flexible and easier to manage:
-
Smarter Cluster Management: Enjoy support for multiple deployment modes, clusters with different worker node flavours via node groups, scalable master nodes and upgraded cluster APIs.
-
Deployment Modes Your Way: Choose between Default Deployment (with bastion host + public load balancer) or Standard Deployment (minimal setup with master + worker nodes).
-
Node Groups for Flexibility: Assign different worker node flavours to specific groups within the same cluster for optimised performance.
-
Node Group APIs: New endpoints let you create, retrieve and delete node groups with ease.
-
High Availability by Default: All clusters are now provisioned with multiple master nodes to improve availability. Plus, you can scale the number of master nodes in a cluster.
New Kubernetes Versions
Hyperstack now supports Kubernetes versions 1.27.8, 1.32.8 and 1.33.4. When deploying a new cluster, simply select your preferred Kubernetes version and Hyperstack will automatically provision the cluster with the compatible image for that version.
Smarter Cluster Deployment
Cluster creation is now even easier: simply choose your preferred Kubernetes version and Hyperstack will automatically select the optimal image and configuration. This streamlines deployment, reduces manual steps and ensures every cluster launches with best-practice settings.
Manual Cluster Reconciliation
You can now manually trigger reconciliation for any Kubernetes cluster. Use the new Reconcile Cluster API endpoint to initiate reconciliation programmatically and streamline cluster management workflows. Explore full details in the API Reference here.
Faster Cluster Creation
Cluster creation and node/group scaling are now faster, so you can deploy and expand workloads with improved speed and efficiency
API Updates for Better Automation
- The List Cluster Versions API now provides detailed version metadata, including supported regions and pre-configured images give your DevOps team greater visibility and control for automated deployments.
- You no longer need to specify an image name when using the Create Cluster API; simply select a Kubernetes version and provisioning will be handled automatically.
Latest Fixes and Improvements
We’ve made several improvements to Hyperstack this month:
- Two-factor authentication setup is now streamlined: simply enter your password, no email address required.
-
Volumes API now shows an array of attachments with details of all associated VM connections.
-
Create Cluster API Enhancements included deployment mode, multi-master setups and node groups right from payload fields.
-
Cluster API response objects now provide detailed
node_groupsandnodesdata. -
Region field now auto-fills when creating environments during VM deployment, reducing errors and saving time.
-
When a volume is protected, it now shows a lock icon with a tooltip on attached virtual machines. You can view the protection status directly through the Volume Attachment APIs.
Important IP Update
A key update regarding hibernating VMs will take effect at the end of this month. Please review the email we sent on Friday for details and look out for additional API instructions in the coming weeks.
If you haven’t subscribed to our newsletter already, please do so to stay updated on all important changes.
New on our Blog
Check out our latest blog on Hyperstack:
Fine-Tuning LLMs for Agentic Use with Hyperstack AI Studio:
A Comprehensive Guide
Building robust agentic systems requires more than simply linking an LLM to tools. Without careful optimisation, tool integrations can expand context windows, increase costs and cause inaccuracies. Fine-tuning specifically for tool use delivers reliable, efficient performance. In this guide, we demonstrate how a support agent which is fine-tuned for integrated tool use can efficiently assist users with tasks like checking fine-tuning job status, accessing documentation and creating support tickets.
Learn more in our latest tutorial.
How to Prepare a Dataset for Fine-tuning on Hyperstack AI Studio:
A Quick Guide
Hyperstack AI Studio is a powerful end-to-end Gen AI platform that simplifies every step of the AI workflow. But before you begin, it’s important to prepare your dataset correctly. Even the most advanced AI models are only as good as the data they are trained on. And poorly formatted data can cause errors, slow training and reduce model performance. In this blog, we’ll take you step by step through preparing your dataset for fine-tuning on Hyperstack AI Studio for a smooth and successful workflow.
Learn more in our latest blog!
Best Cloud GPUs for ComfyUI in 2025:
Power Your AI Workflows like a Pro
ComfyUI is an open-source, node-based program designed for image generation using text prompts. It builds on free diffusion models such as Stable Diffusion and integrates tools like ControlNet and LoRA (Low-Rank Adaptation) to improve functionality. In ComfyUI, each model or tool is represented as a node, helping users map out their AI workflows visually. Our latest blog talks about the best cloud GPUs you will need to run AI workloads with ComfyUI.
Learn more in our latest blog.
How AI Studio Helps You Go from Model to Market Faster:
What You Need to Know
Building an AI product is not just about the model you choose. It is more about the speed with which you can go from idea to a market-ready product. Every extra day you spend on managing the infrastructure or debugging deployment pipelines is a day your competitor could be pitching the same idea to investors or might have already launched the product. In our latest blog, we explore how the Hyperstack AI Studio helps you go from model to market faster.
Learn more in our latest blog.
How to Self-Host n8n on Cloud GPU Server:
A Quick Guide
n8n is an open-source, fair-code workflow automation tool that helps you connect applications, services and APIs through a visual, node-based interface. It offers a simple drag-and-drop automation with the flexibility of JavaScript for advanced customisation. In our quick tutorial, we explore the steps to self-host n8n on a cloud GPU server.
Learn more in our latest blog.
Have an idea you'd like to see in Hyperstack? Let’s bring it to life.
At Hyperstack, we’re committed to continuous improvement and your ideas are a key driver of our innovation.
→ Is there a feature you’ve been waiting for?
→ Something that could speed up your workday?
→ Or a tweak that would make things feel effortless?
Tell us what would make your Hyperstack experience even better. Your feedback sets the direction for what we build next.
Missed the Previous Updates?
Catch up on everything you need to know from Hyperstack Monthly below:
For any questions or suggestions, feel free to reach out at support@hyperstack.cloud. Stay tuned for even more updates and exciting tools next month.
Subscribe to Hyperstack!
Enter your email to get updates to your inbox every week
