TUI Wizard

The TUI wizard guides you through macOS VM creation in 6 steps. Launch it by running the install script on your Proxmox host:

bash -c "$(curl -fsSL https://raw.githubusercontent.com/lucid-fabrics/osx-proxmox-next/main/install.sh)"

Step 1: Preflight Checks

Automatically validates your Proxmox host before proceeding.

Check	What It Verifies
CPU vendor	Intel or AMD detected, applies correct config
Virtualization	VT-x / AMD-V enabled
Root access	Running as root on Proxmox
Dependencies	Required packages installed
TSC flags	constant_tsc and nonstop_tsc for clock stability

No user input required. The wizard auto-detects your hardware and moves forward.

note

Xeon and pre-Skylake Intel CPUs are detected automatically. Xeon keeps -cpu host, older consumer Intel gets Penryn mode with e1000 NIC.

Step 2: Choose macOS Version

Select your target macOS version from the list.

macOS	Apple Services	Notes
Ventura 13	Works	Lightweight, good for older hardware
Sonoma 14	Works	Best tested, most reliable
Sequoia 15	Works (with --apple-services)	Kernel patch applied automatically
Tahoe 26	Works (with --apple-services)	Kernel patch applied automatically

SMBIOS identity (serial, UUID, model) is auto-generated when you select a version.

Step 3: Select Storage

Choose a storage target from auto-detected Proxmox storage pools. The wizard scans for available pools (e.g., local-lvm, NAS mounts under /mnt/pve/).

Step 4: VM Configuration

Review and edit VM settings with auto-filled defaults based on your hardware.

Field	Default	Description
VMID	Next available	Must be unique, 100-999999
Name	macos-{version}	VM display name
CPU Cores	Auto-detected	Must be power of 2 (2, 4, 8, 16)
Memory (MB)	Auto-detected	Minimum 4096 MB
Disk (GB)	Varies by version (80-160)	Minimum 64 GB
Network Bridge	vmbr0	Proxmox bridge interface

Additional options available in this step:

• Generate SMBIOS -- regenerate identity values
• Enable Apple Services -- adds vmgenid, static MAC, and PlatformInfo patching
• Verbose Boot -- shows kernel log instead of Apple logo
• Existing UUID -- enter a UUID to preserve identity for re-runs

warning

macOS requires power-of-2 CPU core counts. Non-power-of-2 values (6, 12) can cause the kernel to hang at the Apple logo.

Step 5: Dry Run

The wizard auto-downloads any missing assets (OpenCore ISO, recovery image), then previews every qm command that will be executed.

This step shows:

• Each command with its title and risk level
• The full qm create, qm set, and qm importdisk sequence
• SMBIOS values that will be applied
• Boot order configuration

Nothing is executed yet. Review the commands before proceeding.

note

Smart caching: OpenCore and recovery images are downloaded once and reused. Creating a second VM with the same macOS version skips the download entirely.

Step 6: Install

Executes all commands from the dry-run preview:

1. Creates the VM with qm create
2. Builds the OpenCore bootloader (GPT + EFI partition)
3. Imports OpenCore and recovery disks via qm importdisk
4. Configures boot order (ide2;virtio0;ide0)
5. Starts the VM

After completion, open the VM console via Proxmox web UI (noVNC) to continue macOS installation.

Post-Install

After macOS finishes installing, fix the boot order so the main disk boots first:

qm set <vmid> --boot order=virtio0;ide0

---

Manage Mode

After creating VMs, launch the wizard with the --manage flag to manage existing macOS VMs:

osx-next --manage

Manage mode lists all macOS VMs detected on the host (identified by isa-applesmc in their config) and provides three actions per VM:

Action	What It Does
Edit	Opens an edit form to change cores, memory, name, network bridge, or disk size. Changes are previewed before applying.
Start / Stop	Toggles the VM power state.
Destroy	Stops and destroys the VM. A config snapshot is saved to generated/snapshots/ before destruction. Rollback hints are shown on failure.

The edit form in manage mode behaves the same as the edit subcommand: the VM is stopped before changes are applied, existing MAC address and NIC model are preserved when changing the bridge, and the VM can optionally be restarted after.