How to Install Home Assistant: Every Method Compared

This article covers the same ground as the video below but in a format you can reference while you're at your bench with hardware in hand. If you'd rather watch than read, hit play. If you want the written version with additional context on each method, keep scrolling.

↗ View the full video page

Why Home Assistant is worth the setup cost

Home Assistant is not the easiest smart home platform to get running. That's a deliberate trade-off, not an oversight. The extra friction at setup time is what buys you local control, no recurring subscription fees, deep integration with thousands of devices, and a level of automation logic that cloud-dependent platforms simply cannot match.

Once it's running, the maintenance burden is low. Getting there requires choosing the right installation method for your hardware and comfort level — and that choice has meaningful long-term consequences. Here's how each option stacks up.

Installation method 1: Home Assistant Green

If you have no experience imaging operating systems or working with microcomputers, start here. Home Assistant Green is a dedicated plug-and-play appliance that ships with the full Home Assistant OS pre-installed. There is no imaging step, no assembly, and no configuration before first boot.

The hardware spec is modest but adequate for a standard home automation workload: a 1.8 GHz quad-core ARM processor, 4 GB of RAM, and 32 GB of onboard storage. Setup is genuinely as simple as connecting an Ethernet cable, plugging in the included power adapter, and navigating to the local URL from any browser on your network. At around $100 from the Home Assistant website, it is the lowest barrier to entry of any option on this list.

The trade-off is that the hardware is fixed. You cannot upgrade the storage, swap the compute, or expand the I/O. If you eventually outgrow it — running many add-ons, storing large amounts of historical sensor data — you'll be migrating to a different platform. For most households running a straightforward automation setup, that ceiling is a long way off.

Installation method 2: Raspberry Pi with Home Assistant OS

A Raspberry Pi gives you more hardware flexibility than the Green at roughly comparable cost once you account for accessories. You source the compute and storage separately, which means you can spec to your actual requirements — more RAM, faster storage, a larger SD card or SSD — rather than accepting fixed hardware.

The process requires imaging a micro SD card or SSD with the Home Assistant OS image using Raspberry Pi Imager, then assembling and booting the hardware. It's not technically difficult, but it does require following a sequence of steps correctly. If you're buying new, a starter kit is the practical choice — they include everything required and eliminate compatibility guesswork.

This is the method covered in the full installation walkthrough later in this article.

Installation method 3: Home Assistant Yellow

Home Assistant Yellow is a purpose-built carrier board and enclosure that requires you to supply a Raspberry Pi Compute Module. It sits between the Green (fully assembled) and a bare Raspberry Pi (fully DIY) in terms of assembly effort.

What it adds over both: an M.2 slot for SSD expansion, an optional PoE port so it can draw power from a PoE-capable network switch rather than a wall adapter, and a built-in Zigbee radio with Matter compatibility. For anyone planning to run Zigbee devices directly from the Home Assistant host without a separate Zigbee coordinator dongle, this is worth considering seriously. PoE also simplifies cable management if your equipment closet already has a managed PoE switch.

Installation method 4: Generic x86 hardware, VMs, and containers

Home Assistant runs on standard 64-bit x86 hardware because it's Linux underneath. If you have a spare machine sitting idle, this is a viable path to get started without buying dedicated hardware.

Virtual machine and Docker container installs introduce a specific set of limitations worth understanding before you commit. When you run Home Assistant inside a VM or container rather than directly on hardware, certain add-ons may not function correctly, and the automatic update mechanism that the Supervisor provides in a full OS install may behave differently. USB device passthrough (for Zigbee dongles, Z-Wave sticks, and similar hardware) requires extra configuration and occasionally breaks across host updates.

Use this method if you need to run Home Assistant alongside other workloads on existing infrastructure, or if you want to evaluate it before committing dedicated hardware. Go in knowing the limitations and you won't be caught off guard.

What we're installing: Raspberry Pi 5 with Home Assistant OS

For the remainder of this article, we're doing a full install of Home Assistant OS on a Raspberry Pi 5. This gives you the full feature set — Supervisor, add-ons, OTA updates — with real compute headroom and no meaningful constraints for a standard home automation workload.

The kit used here is a Raspberry Pi 5 starter kit from Canakit. It includes a Raspberry Pi 5 with a 2.4 GHz 64-bit quad-core CPU and 8 GB of RAM, a 128 GB micro SD card, a heat sink, a cooling fan, display cables, an enclosure, and a USB-C power supply. Home Assistant doesn't need 8 GB of RAM for basic operation — this is the available kit, not a minimum requirement.

Storage: micro SD vs. SSD

A micro SD card works and will get you started. Two things to know before you rely on one long-term.

First, use an A2-rated card. A2 is the application performance class rating for SD cards, meaning the card is designed to handle the kind of small, frequent read/write operations that an application like Home Assistant generates continuously. A standard card rated for sequential reads and writes will degrade significantly faster under this workload.

Second, Home Assistant logs aggressively by default. That constant write activity will eventually wear out any SD card. For a permanent installation, a solid-state drive connected via USB is a substantially more durable choice. SD card to start, SSD for production — that's the practical sequence.

Home Assistant requires a minimum of 16 GB of storage. 32 GB is a reasonable baseline for a real setup with a few add-ons and some historical data retention.

Step 1: Image the SD card

Download and install Raspberry Pi Imager on your computer. Insert the micro SD card into the USB card reader included with the kit, then plug the reader into your computer.

Launch Raspberry Pi Imager and work through these three selections in order:

1. Device: Select Raspberry Pi 5.
2. Operating System: Navigate to Other specific-purpose OS → Home assistants and home automation → Home Assistant → Home Assistant OS.
3. Storage: Select the SD card you plugged in. If the Canakit card shipped with Raspberry Pi OS pre-installed, it is safe to overwrite — that's the point of this step.

Click Next and wait for the image to write and verify. Once complete, eject the card reader and remove the SD card. Do not skip the verify step — a failed write that passes verification is far less common than a failed write you'll only discover after assembly when the Pi won't boot.

Step 2: Assemble the hardware

The Canakit enclosure assembly is straightforward. Here's the sequence:

1. Separate the enclosure into its three parts: base plate, housing, and top plate.
2. Angle the Raspberry Pi board slightly and seat it onto the base plate. The USB ports align with the cutouts on the base plate — use that as your alignment reference. It only fits one way.
3. Install the heat sink. It ships with thermal pads pre-attached. One pad is oriented differently from the others — that one goes on the Ethernet transceiver chip. Remove the protective film from all pads, align the heat sink over the chips, and press down firmly.
4. Before fitting the housing, locate the fan header next to the GPIO interface and remove the cover. Connect the cooling fan — it only inserts one way, with the yellow wire facing away from the board.
5. Place the housing over the board, aligning the rear USB and Ethernet ports with the housing cutouts. Route the fan cable up through the top opening.
6. Attach the cooling fan to the top plate with the logo facing outward, then secure the top plate to the housing.
7. Flip the case over and insert the imaged SD card into the slot on the underside, logo facing outward.

Power supply note

The Raspberry Pi 5 draws more power than previous Pi generations, particularly when powering USB peripherals. Use only the included power supply or a verified equivalent rated for the Pi 5. Underpowering the board produces inconsistent behavior that is frustrating to diagnose — it looks like software problems when the cause is hardware.

Step 3: First boot and initial setup

Connect the Ethernet cable and plug in the power supply. Give the Pi a minute or two to complete its first boot — it's initializing the OS and running first-time setup tasks, so it takes slightly longer than subsequent boots.

From any other device on the same network, open a browser and navigate to homeassistant.local:8123. The Home Assistant mobile app also works if you prefer that interface.

If you can't connect

If the hostname doesn't resolve, the Pi has an IP address even if mDNS isn't working on your network. A few ways to find it:

• Install the Fing app on your phone and run a network scan. The Pi will appear in the device list.
• Check your router's DHCP client table — look for a device named homeassistant.
• Connect a monitor via the micro HDMI adapter and a USB keyboard, then open the terminal and run ifconfig to see the assigned address.

If none of those produce a result, the image likely didn't write correctly. Re-image the SD card and try again. This is almost always the cause when a Pi won't boot to Home Assistant at all.

Step 4: Create your account and configure your home

The welcome screen may show a "Preparing Home Assistant" message while the system finishes initializing. Once the setup wizard appears:

1. Create My Smart Home — this starts the onboarding flow.
2. Create a user account. Choose a strong password and store it in a password manager — not your browser. Home Assistant has no password recovery mechanism. If you lose the admin password, account recovery requires direct filesystem access to the running system.
3. Set your location. This creates your Home zone — a geographic boundary representing your home address. The Home zone is used in any automation involving presence detection or location-based triggers. It also sets your units of measurement, time zone, and local sunrise/sunset times, which several built-in automations rely on.
4. Telemetry preferences. Home Assistant asks whether to share anonymized usage data. The data is used to prioritize development effort. You are not required to share anything — the system works identically either way.

After setup completes, you'll land on the main Home Assistant dashboard. Anything Home Assistant detected on your network during onboarding will be listed for review in Settings. From here, the next step is adding integrations for your devices, building your first automations, and configuring a dashboard that reflects how your home actually works.

What to do next

You now have a running Home Assistant instance on solid hardware with the full feature set available. A few things worth doing before you go further:

• Take a backup immediately. Settings → System → Backups. Before you configure anything you'll have to redo, create a baseline backup. Store it somewhere other than the Pi itself.
• Check for updates. Your installed image may not be the latest version. Settings → System → Updates. Apply any pending updates before adding devices.
• Review discovered devices. Settings → Devices & Services. Home Assistant scans your local network during setup and may have already found devices. Review what it found before manually adding integrations.
• Plan before you click. The integrations library has thousands of entries. Inventory the smart devices you already own, look up each one in the Home Assistant integration directory, and understand what each provides before connecting anything.

The companion article on getting started with Home Assistant covers the core concepts — entities, devices, integrations, automations — in depth. That's the right next read once the hardware is running.