TABLE OF CONTENTS
NVIDIA H100 SXM On-Demand
We’re back with your monthly dose of Hyperstack updates!
November is packed with exciting features on Hyperstack. From AI Studio updates to VM IP Update on Hibernation, there is so much to explore.
Even more exciting is that we’re bringing you a line-up of interesting blogs straight from our team of experts, you won’t want to miss.
New on Hyperstack
Check out what's new on the Hyperstack this month:
Public IP Behaviour Change During Hibernation
You can now choose to keep your VM’s public IP address during hibernation. By default, hibernation will automatically release the public IP, helping to minimise idle resource costs. Learn how to hibernate a Virtual Machine using the UI.
New on AI Studio
Here’s what’s new on our full-stack Gen AI platform, AI Studio this month:
- Import LoRA Adapters Directly From Hugging Face: No more manual downloads or messy workflows. You can now import external LoRA adapters from Hugging Face straight into AI Studio and plug them into supported base models. Use them instantly in the AI Studio Playground or deploy them via API for inference.
- Sample Datasets Now in the UI: No more hunting for starter data. You’ll now find a curated sample dataset directly inside the AI Studio interface, perfect for quick experiments, testing or getting hands-on without setup friction.
- Export Your Fine-Tuned Models: You can now export any fine-tuned model you create in AI Studio and use it for external use, giving you more control and flexibility in your ML workflows.
New Deployment Coming Live
A new deployment for NVIDIA RTX Pro 6000 SE will be going live soon.
The NVIDIA RTX Pro 6000 Blackwell provides an instant uplift to the workloads you already run every day. If your workloads are consistent, securing access early to our new deployment ensures you stay ahead of the curve. Reservation starts at just $1.26/hr.
Fixes and Improvements
- Networking is now more reliable when deploying multiple VMs in a new environment. The first VM manages the setup to avoid any configuration conflicts.
- A new
retain_ipparameter has been added to the Hibernate VM API, so you can programmatically decide whether your VM's public IP stays attached during hibernation.
Coming Soon on Hyperstack
Here’s a glimpse of what’s coming next:
- Object Storage: Get ready for Hyperstack Object Storage, a scalable storage solution to handle unstructured data on any scale. Built on Amazon S3-compatible technology, it delivers a secure, flexible and API-ready way to manage everything from AI/ML datasets to backups and media files.
Stay tuned for the latest updates in our upcoming rundowns.
New on the Blog: Insights From Our Team
This month is especially exciting on our blog. Our team is sharing some truly insightful pieces and you’ll definitely want to give them a read:
Run DeepSeek OCR on Hyperstack with your Own UI:
A Comprehensive Guide
In our latest tutorial, we show how to set up DeepSeek-OCR on a Hyperstack VM to create a high-performance, private OCR workflow. DeepSeek-OCR is a 3-billion-parameter multimodal model that combines a vision encoder and language decoder to extract text and preserve document structure, including tables and complex layouts. Using vLLM for GPU-accelerated serving, you can run PDFs and images through a simple Gradio UI or build a custom REST API.
Learn more in our latest tutorial.
ECC on NVIDIA H100 PCIe VMs:
How to Enable or Disable It and Why It Matters
In our latest tutorial, we explain the importance of ECC (Error-Correcting Code) on NVIDIA H100 PCIe VMs. ECC detects and corrects memory errors, ensuring data integrity for AI, HPC, and scientific workloads. We guide you through checking, enabling or disabling ECC safely on Hyperstack for balancing performance and reliability.
Learn more in our latest tutorial.
Become an Expert with Prometheus on Hyperstack:
Tips and Tricks
In our latest blog, we show how to set up Prometheus on your own Hyperstack VM to gain full control over monitoring your infrastructure. Prometheus collects, stores, and analyses metrics from VMs, containers, and applications, enabling alerts for issues like high CPU usage, slow applications, or downtime. We guide you through installing Docker, configuring Prometheus, monitoring additional servers with Node Exporter and applying zero-downtime config updates.
Learn more in our latest blog.
Deploying Mistral Large 3 on Hyperstack:
A Step-by-Step Guide
Mistral has released Mistral Large 3, its most powerful multimodal model with frontier-level performance across reasoning, vision and agentic workflows. Whether you're building advanced assistants, enterprise automation or large-scale multimodal applications, Mistral Large 3 delivers the reliability and capability you need. Explore our full tutorial below to deploy and run Mistral Large 3 with ease.
Learn more in our latest tutorial.
How to Integrate Hyperstack AI Studio with Claude Code:
A Beginner's Guide
To power Claude Code with custom models, we need a robust, flexible AI backend that can serve models, support fine-tuning, and provide scalable inference. This is where Hyperstack AI Studio becomes crucial. Hyperstack provides a complete generative AI platform where developers can host, fine-tune, benchmark, and deploy LLMs with OpenAI-compatible APIs, making it fully compatible with tools like Claude Code.
Learn more in our latest tutorial.
Your Ideas Power Hyperstack
You know your workflow better than anyone. If there’s anything you wish Hyperstack did differently or better, now’s your chance to tell us.
Maybe it’s a feature you’ve been thinking about, a tool that could speed up your workflow, or a simple improvement that would make your project easier. Whatever it is, we’re listening.
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
