Proxmox VE 9.2 Arm64: Official Linux Virtualization on ARM Servers — Install Guide

On August 5, 2026, Proxmox Server Solutions made an announcement that homelab and enterprise communities had been requesting since at least 2019. Proxmox VE Arm64 Linux server support is now officially here — not a community port, not a beta, and not a tech preview. Proxmox VE 9.2 for Arm64 shares the exact same codebase, package repositories, release lifecycle, web interface, and configuration tools as the x86-64 version. Built on Debian 13.5 Trixie with Linux kernel 7.0, it ships QEMU 11.0, LXC 7.0, and ZFS 2.4. After more than two decades of x86 exclusivity, the most popular open-source hypervisor just became an architecture-agnostic platform. This guide covers exactly what is included, what the limitations are, which hardware is supported, and how to install and configure Proxmox VE 9.2 on an Arm64 server today.

What Is Proxmox VE 9.2 Arm64 and Why It Matters

Proxmox VE is an open-source virtualization platform that combines KVM hypervisor and LXC container support in a single web-managed environment. Because it runs on bare metal and provides enterprise features at zero licensing cost, it has become the dominant self-hosted hypervisor for both homelabs and production data centers. Until now, it ran only on x86-64 hardware — a limitation that excluded the rapidly growing Arm server market entirely.

That changes with 9.2. Since the Arm64 build pulls from the same repositories as x86-64, you get automatic updates through the same channels, the same security patches on the same schedule, and the same paid enterprise repository for production support. For data center operators evaluating Arm-based servers for their energy efficiency and high core-count density, this is significant: you can now run your proven Proxmox workflows on Ampere Altra, NVIDIA Grace Hopper, and AWS Graviton-class hardware without any community-maintained detours.

Full Component Stack on Arm64

The component versions in Proxmox VE 9.2 Arm64 are identical to the x86-64 release. Here is what ships out of the box:

Supported Hardware and Key Requirements

Proxmox draws a clear line between fully supported silicon and best-effort hardware. Understanding this distinction before purchasing hardware saves significant troubleshooting time.

Fully Supported Platforms

Proxmox provides full first-party support and validation on two NVIDIA platforms:

• NVIDIA Grace Hopper — the Grace CPU + Hopper GPU superchip used in DGX GH200 and HGX GH200 systems
• NVIDIA Vera Rubin — the next-generation Vera CPU + Rubin GPU platform for AI factory infrastructure

Best-Effort Support

Beyond the NVIDIA platforms, Proxmox extends best-effort support to:

• Any UEFI-based ARMv9-A server hardware (boots via UEFI, describes hardware via ACPI)
• ARMv8-A systems that meet the same UEFI + ACPI requirement
• Ampere Altra and Altra Max systems with UEFI firmware
• AWS Graviton instances running on bare metal EC2 instance types

Not Supported

• Device-tree-only SBCs such as Raspberry Pi — the host must boot via UEFI and use ACPI
• SeaBIOS legacy BIOS boot — AAVMF UEFI is required for VMs
• AMD SEV encrypted memory — not available on Arm64 in this release

# Verify your Arm64 hardware boots via UEFI before installing
# Check UEFI boot from a live environment
ls /sys/firmware/efi 2>/dev/null && echo 'UEFI boot confirmed' || echo 'Not UEFI -- unsupported'

# Check ACPI support
ls /sys/firmware/acpi 2>/dev/null && echo 'ACPI present' || echo 'No ACPI -- likely unsupported'

How to Install Proxmox VE 9.2 on Arm64

Installation follows the same process as x86-64, using a dedicated ISO for Arm64. Because the ISOs are architecture-specific, make sure you download the Arm64 ISO and not the x86-64 version.

Option 1: Bare Metal ISO Install

# Download the Proxmox VE 9.2 Arm64 ISO
# From the Proxmox downloads page: https://www.proxmox.com/en/downloads
# Select: Proxmox VE 9.2 ISO Installer (arm64)

# Write to USB drive (replace /dev/sdX with your USB device)
dd if=proxmox-ve_9.2-1_arm64.iso of=/dev/sdX bs=1M status=progress
sync

# Boot the target Arm64 server from the USB drive
# The installer is identical to the x86-64 version
# Select target disk, set timezone, configure network, set root password

Option 2: Add Arm64 Repository to Existing Debian 13.5 Trixie

If you already have Debian 13.5 running on an Arm64 server, you can add the Proxmox VE repository and install without a full reinstall:

# Confirm you are on Debian 13.5 Trixie arm64
cat /etc/debian_version
dpkg --print-architecture
# Must return: arm64

# Add the Proxmox VE no-subscription repository
echo 'deb [arch=arm64] http://download.proxmox.com/debian/pve trixie pve-no-subscription' \
  > /etc/apt/sources.list.d/pve-install-repo.list

# Add the Proxmox repository key
curl -fsSL https://enterprise.proxmox.com/debian/proxmox-release-trixie.gpg \
  -o /etc/apt/trusted.gpg.d/proxmox-release-trixie.gpg

# Install Proxmox VE
apt update && apt full-upgrade -y
apt install proxmox-ve postfix open-iscsi chrony -y

# Remove Debian default kernel (Proxmox kernel replaces it)
apt remove linux-image-arm64 'linux-image-6*' -y
update-grub

# Reboot into Proxmox VE kernel
reboot

Post-Install Verification

After the reboot, verify that Proxmox VE is running correctly on the Arm64 host before adding storage or creating any VMs:

# Confirm Proxmox kernel is running
uname -r
# Should show: 7.0.x-x-pve-arm64 or similar

# Confirm architecture
uname -m
# Should return: aarch64

# Check Proxmox VE version
pveversion
# Should show: pve-manager/9.2.x/...

# Verify all Proxmox services are running
systemctl is-active pvedaemon pveproxy pvestatd pvescheduler

# Access the web UI
# https://YOUR_SERVER_IP:8006
# Login: root / your-root-password

# Check cluster status (even single node shows as healthy)
pvecm status 2>/dev/null || echo 'Single node -- no cluster configured'

Configure ZFS Storage on Arm64

ZFS 2.4 on Arm64 behaves identically to the x86-64 version. Since ZFS is one of the primary reasons many sysadmins choose Proxmox, verifying it works correctly after install is an important first step:

# List available disks for ZFS pool creation
lsblk
fdisk -l | grep 'Disk /dev'

# Create a ZFS mirror pool (replace sdb and sdc with your disk names)
zpool create -f -o ashift=12 \
  -O compression=lz4 \
  -O atime=off \
  datastore mirror /dev/sdb /dev/sdc

# Verify pool health
zpool status datastore
zpool list

# Add ZFS pool to Proxmox storage
pvesm add zfspool datastore-zfs \
  --pool datastore \
  --content images,rootdir

# Verify storage appears in Proxmox
pvesm status

For expanding ZFS pools on Arm64 as your storage needs grow, the same LVM and ZFS expansion commands apply as on x86-64. Our LVM disk management guide covers volume expansion techniques that complement ZFS pool management on Proxmox hosts.

Running Mixed x86-64 and Arm64 Clusters

One of the most practical questions for existing Proxmox users is whether Arm64 nodes can join an existing x86-64 cluster. The answer is yes — with an important caveat about migration.

Since Proxmox VE 9.2 Arm64 and x86-64 share the same cluster protocol, both architectures can participate in the same cluster for management purposes. You get unified monitoring, shared firewall rules, and shared SDN configuration across all nodes. However, live migration is restricted to same-architecture pairs. An Arm64 VM cannot live-migrate to an x86-64 node because the CPU instruction sets are incompatible. Cold migration between architectures is possible if the guest OS supports both Arm64 and x86-64 — which most Linux distributions do, since they publish both architecture builds.

# Add an Arm64 node to an existing cluster
# Run on the Arm64 node after it has Proxmox VE 9.2 installed

# First, get the join information from the existing cluster master
# On the MASTER node:
pvecm status
pvecm nodes

# Join the cluster from the Arm64 node
# Replace MASTER_IP with the IP of an existing cluster node
pvecm add MASTER_IP

# Verify the Arm64 node joined successfully
pvecm status
pvecm nodes

# The Arm64 node should appear in the web UI under Datacenter

Creating and Running VMs on Arm64

Creating VMs on the Arm64 edition follows the same web UI workflow as x86-64, but with one key difference: VMs must use AAVMF UEFI firmware instead of SeaBIOS. Because of that requirement, legacy BIOS-only operating system images will not boot. Modern Linux distributions all support UEFI boot, so this is rarely a practical limitation:

# Create an Arm64 VM via CLI (pvesh or qm)
# Ubuntu 26.04 LTS arm64 example

# Download Ubuntu 26.04 arm64 cloud image
cd /var/lib/vz/template/iso/
wget https://cloud-images.ubuntu.com/releases/26.04/release/ubuntu-26.04-server-cloudimg-arm64.img

# Create VM with UEFI firmware (required for Arm64)
qm create 100 \
  --name ubuntu-arm64 \
  --memory 4096 \
  --cores 4 \
  --net0 virtio,bridge=vmbr0 \
  --bios ovmf \
  --machine virt \
  --efidisk0 datastore-zfs:0,format=raw

# Import the disk image
qm importdisk 100 ubuntu-26.04-server-cloudimg-arm64.img datastore-zfs

# Attach the imported disk and set boot order
qm set 100 \
  --virtio0 datastore-zfs:vm-100-disk-1 \
  --boot order=virtio0

# Start the VM
qm start 100
qm status 100

Known Limitations on Arm64 vs x86-64

Before migrating production workloads, understand the current limitations of the Arm64 release compared to the x86-64 version:

Security Hardening After Installation

The post-install security baseline for Proxmox VE Arm64 is identical to x86-64. Because Proxmox exposes a web management interface and an API on port 8006, hardening that surface is essential before putting the host into production:

# Restrict web UI access to management network only
# Edit /etc/pve/user.cfg to review current user permissions
cat /etc/pve/user.cfg

# Change default SSH port and disable password auth
# Edit /etc/ssh/sshd_config
grep -E 'PasswordAuthentication|PermitRootLogin|Port' /etc/ssh/sshd_config

# Set up firewall for the Proxmox host
# Enable host firewall in web UI: Datacenter > Firewall > Options
# Or via CLI:
pvesh set /nodes/$(hostname)/firewall/options --enable 1

# Add rule to allow only management IP to port 8006
pvesh create /nodes/$(hostname)/firewall/rules \
  --action ACCEPT \
  --type in \
  --proto tcp \
  --dport 8006 \
  --source YOUR_MGMT_IP

# Verify firewall status
pvesh get /nodes/$(hostname)/firewall/options

Our Linux server hardening checklist covers the broader OS-level hardening baseline that applies to Proxmox VE hosts on both architectures — including SSH key configuration, audit logging, and kernel parameter tuning. The official Proxmox Arm64 release announcement is at proxmox.com, and a community-focused technical breakdown is at linuxiac.com.

Conclusion

After more than two decades of x86-64 exclusivity, Proxmox VE Arm64 Linux server support is now a production reality. The 9.2 release delivers full feature parity for KVM, LXC, ZFS, Ceph, clustering, and enterprise support on Arm64 hardware. If your target hardware boots via UEFI with ACPI, start with the repository-based installation on Debian 13.5 Trixie — it is the least disruptive path for existing Linux deployments. After the install, verify the Proxmox kernel is running with uname -r, check all services are active, and set up your ZFS storage pool before creating any VMs. For mixed-architecture clusters, remember that live migration stays within architecture boundaries — plan your VM placement accordingly from day one. The energy efficiency and core density advantages of Arm64 servers make this release particularly relevant for data centers evaluating their infrastructure costs. Proxmox on Arm64 is no longer a community experiment. It is a supported, production-grade platform.