---
title: "Unikraft Cloud Blog"
description: "Latest posts and technical articles from Unikraft Cloud."
doc_version: "1.0"
last_updated: "2026-11-13"
---

# Unikraft Cloud Blog

## [Live Migration, now live at Unikraft](https://unikraft.com/blog/live-vm-migration)

Unikraft now live-migrates microVMs (and their volumes) between nodes while they keep running. This means your placement decisions can be made and adapted on the fly, depending on hardware, workload, or cost.

## [Release 15 “Isonoe”: Live Migration, Windows VMs, Terminal in the Dashboard and More](https://unikraft.com/blog/isonoe-15-release)

MicroVMs are no longer tied to the node they started on, Windows VMs join the platform, and yet another big set of improvements to the dashboard.

## [Panic at the Dinner Table: Which Fork Do I Use??](https://unikraft.com/blog/branch-is-not-fork)

A few days ago, I was nerd sniped in a customer call with the definition of fork. Since I am a systems engineer by trade, I felt like the record needed to be set straight.

## [Release 14 “Harpalyke”: Vertical Scaling, GPU Snapshots and Sandboxes](https://unikraft.com/blog/harpalyke-14-release)

Instances that resize while they run, GPUs that are freed whenever an instance is idle, sandboxes in the CLI, dashboard and JavaScript SDK, and Dockerfile builds. Here is everything that's new in R14.

## [Say No to Idle GPUs](https://unikraft.com/blog/no-idle-gpus)

The industry has spent three years buying GPUs and almost no time using them well. Up to 70% of a fleet can sit idle because a card stays pinned to an instance that is doing nothing. Here is why an idle GPU should belong to someone else — and how GPU snapshots make that possible.

## [How Netlify Runs all of Edge Functions on Unikraft microVMs](https://unikraft.com/blog/netlify-edge-functions)

Netlify now runs every tier of Edge Functions on Unikraft microVMs on their own infrastructure. This allows them to cut median per-request overhead from roughly 50ms to 6ms, while keeping full control of the runtime.

## [Release 13 “Adrastea”: ARM64, Kernel-less Images and a JavaScript SDK](https://unikraft.com/blog/adrastea-13-release)

ARM64 hosts and multi-platform builds, kernel-less images, elastic Kubernetes capacity through Karpenter, a JavaScript SDK, flexible networking, and a substantially expanded dashboard. Adrastea is the biggest release since we shipped GPUs.

## [Millisecond, microVM Kubernetes](https://unikraft.com/blog/millisecond-microvm-kubernetes)

How we brought microVM-level isolation and speed to Kubernetes: the architectural journey behind Kraftlet, the ideas we explored, the dead ends we hit, and how you can run it on your existing cluster.

## [Release 12 “Thebe”: GPUs and Better Image Control](https://unikraft.com/blog/thebe-12-release)

NVIDIA GPU support comes to the platform, instance templates go cluster-wide, and the dashboard gets a big update. Here is everything that's new in R12.

## [The Ten-Millisecond Agent](https://unikraft.com/blog/ai-sandboxes-at-scale)

How to run AI sandboxes at scale and (almost more importantly) how not to. I'll go into the forces that break agent infrastructure, and the platform we built from first principles to survive them: Millions of agents, ~10 ms cold starts, stateful scale-to-zero, and a strong hardware boundary around every one.

## [The Great Fragmentation of Cloud Compute](https://unikraft.com/blog/cloud-compute-fragmentation)

Why does a single cloud sell you four compute services for one workload — and what would it take to defragment them back into one?

## [The Big AI Divide](https://unikraft.com/blog/big-cloud-divide)

AI collapsed the cost of building a virtualization platform, but running one in production at scale is just as hard as ever. Here's where the real work now lives.

## [Say No to Warm Pools](https://unikraft.com/blog/say-no-to-warm-pools)

You provision always-on clusters, provisioned for a spike that spends most of its life not happening. Here is what they actually cost you — and why millisecond cold starts make the whole pattern obsolete.

## [Why do I need Unikraft? I'll just grab Firecracker](https://unikraft.com/blog/unikraft-vs-firecracker)

Firecracker is a superb virtual machine monitor, but it is only one component of a full platform. Here is exactly what you get when you run Unikraft instead of building it yourself.

## [Release 11 “Metis”: Snapshots, Sandboxes, and a Shield](https://unikraft.com/blog/metis-11-release)

Branching and checkpointing, persistent custom filesystems, a plugin API, and a network shield that keeps secrets out of your microVMs entirely — plus a completely revamped dashboard. Here is everything in R11.

## [The Mighty (micro)Virtual Machine: The Only Isolation Primitive That Actually Isolates](https://unikraft.com/blog/the-mighty-microvm)

A no-BS guide to VMs, containers, microVMs, and isolates — and why, for multi-tenant workloads, there is, and only ever was, one right answer.

## [Cramming 1M (Scaled to Zero) Virtual Machines in a Single Box](https://unikraft.com/blog/1m-vms-single-box)

At Unikraft, we're on a quest for exponentially better cloud efficiency. We redesigned our platform to fit over 1 million scaled-to-zero microVMs on a single standard server, with cold starts and resumes still in single-digit milliseconds.

## [AI Startup? You Need to Think about Scale from Day 1](https://unikraft.com/blog/ai-startup-think-of-scale-day-one)

The era of "move fast, fix infra later" is over for AI companies. It used to be the case that if you were an early stage startup the name of the game would be to get whatever infra would get you started the quickest. That appears to no longer be the case.

## [We Raised $6M to Build the Future of (AI) Cloud Infrastructure](https://unikraft.com/blog/we-raised-6m-to-build-the-future-of-ai-cloud-infra)

Today, and after years of hard work, we’re ecstatic to announce the launch of Unikraft Cloud, a radical, new cloud platform based on years of research and providing exponentially better scalability and efficiency (think running millions of strongly isolated instances on a few servers instead of an entire datacenter). In support of this, we’re also super proud to announce our latest $6M seed raise with Heavybit as lead investor, with the participation of Vercel Ventures, Mango Capital, Firestreak Ventures, Fly VC and First Momentum Ventures, along with a host of world class angels.

## [Running Thousands of Millisecond Headless Browsers on a Single (Cloud) Server](https://unikraft.com/blog/millisecond-headless-browsers)

Headless browsers have become a fundamental cog in the AI machine, running as a cloud service for running JavaScript code, scraping websites, automating testing, and integrating with AI agents. A headless browser is, however, a resource-hungry, heavy beast. In this article we show how Unikraft Cloud tech allows for deploying 1000s of these on a single server, while ensuring they can be started in 10s of milliseconds.

## [SOC2-I Compliance: An Intense but Extremely Valuable Journey](https://unikraft.com/blog/soc2-i-compliant)

At Unikraft, we recently completed our SOC2 Type 1 audit. What we thought would be a process of improving security turned into a much more valuable journey about company maturity.

## [The Idle (and Over-provisioned) Cloud](https://unikraft.com/blog/the-idle-overprovisioned-cloud)

We generally like to think of the cloud handling gazillion requests per second, but a significant part of it is over-provisioned and so idle. In this post we look into the what, how and why what we term "The Idle Cloud" exists, and how to fix it.

## Sitemap

- [Home](https://unikraft.com/)
- [How It Works](https://unikraft.com/how-it-works)
- [AI](https://unikraft.com/ai)
- [Pricing](https://unikraft.com/pricing)
- [Customer Stories](https://unikraft.com/customer-stories)
- [Blog](https://unikraft.com/blog)
- [Documentation](https://unikraft.com/docs)
- [Glossary](https://unikraft.com/glossary.md)
- [Contact](https://unikraft.com/contact)
