Home Assistant + Frigate hardware

Best Hardware for Home Assistant and Frigate

Home Assistant is usually not the reason a combined server needs stronger hardware; Frigate is. Cameras add continuous decoding, object detection, recording writes and bursts of activity, while Home Assistant adds automations, integrations, radios and database work. The best shared server keeps those jobs isolated enough that a busy camera event never makes the smart home feel sluggish.

See live hardware Size my workload

Quick answer

What this combined workload needs

For a modest installation, a modern Intel mini PC with 16GB RAM, hardware video decoding and a supported Frigate detector is a strong baseline. Choose 32GB when Home Assistant has many apps or the host also runs databases, media, semantic search or additional containers.

Live Amazon hardware

Current products that fit this decision

These listings come from this sprint's dedicated Amazon catalogue. Barebones PCs, laptops, GPU enclosures, adapter-only accelerator listings, unrelated accessories and ambiguous radio products are excluded by the normalizer.

Checking the dedicated Home Assistant + Frigate + AI catalogue…

Buying decision

Match the hardware to the bottleneck

Start with camera count and simultaneous activity, then decide the detector. After that, choose how Home Assistant will run and how Zigbee/Thread/Z-Wave devices will reach it. Do not size the machine from Home Assistant minimum requirements and assume Frigate will fit in the leftover capacity.

Interactive decision tool

Home Assistant + Frigate Hardware Sizer

Size the combined workload before comparing products. Results are planning tiers, not benchmark guarantees. Verify current Home Assistant installation support, Frigate detector/device access and Ollama hardware support for the exact system.

Compatibility checklist

Four checks before you purchase

Assign every accelerator a job

Video decoding, Frigate detection, enrichments and Ollama inference are separate workloads. Do not assume one NPU/GPU automatically covers all of them.

Protect Home Assistant headroom

Size the combined host for peak overlap so camera or LLM activity cannot starve automations, databases or radio services.

Verify device access

USB radios, iGPU/NPU/GPU and Hailo devices must be visible to the correct host, container or VM before their benchmark performance matters.

Keep data roles separate

Use fast SSD space for system/models and size Frigate recordings independently so retention cannot crowd the smart-home environment.

Frigate usually defines the compute floor

Home Assistant can run on very modest hardware, but Frigate continuously processes video. The shared server therefore needs enough media-engine capacity, detector throughput and memory for the camera workload before Home Assistant overhead is considered.

This is why a Raspberry Pi that feels fast with automations can become a poor camera server. The upgrade is driven by video and AI, not by dashboards or YAML.

Hardware decoding should be the first Frigate check

A modern Intel iGPU can remove a large amount of video decode work from the CPU. AMD and NVIDIA paths can also work when supported and configured correctly. If ffmpeg falls back to software decoding, CPU usage can stay high even after adding an AI detector.

Before spending on a larger processor, confirm that the decode device is visible inside the Frigate environment. This one configuration detail can change the size of the host you actually need.

Object detection should have its own plan

Frigate supports dedicated accelerators and integrated GPU/NPU paths. Hailo and Intel OpenVINO are attractive new-build directions, while Coral remains useful in some existing or constrained systems.

The detector is chosen from simultaneous activity and model needs, not Home Assistant entity count. A camera pointed at a busy sidewalk can create more inference pressure than several quiet indoor cameras.

Sixteen gigabytes is a sensible shared baseline

A small Frigate instance and Home Assistant can operate with less, but 16GB gives practical room for databases, apps, MQTT, integrations and temporary memory spikes. Frigate recommends more memory as camera/enrichment workloads grow.

Use 32GB if the machine also runs semantic search, face recognition, Node-RED, databases, media services or other containers. Memory headroom is cheap compared with diagnosing swapping on an always-on automation host.

Home Assistant OS and Container have different tradeoffs

Home Assistant currently supports Home Assistant OS and Home Assistant Container, with Home Assistant OS recommended for most users. OS includes the Supervisor/apps experience; Container gives more host control but requires you to manage the Linux/Docker environment and does not include Home Assistant apps.

On a combined appliance, HA OS in a VM gives Home Assistant a clean lifecycle while Frigate runs separately. A Docker-first user may prefer Home Assistant Container beside Frigate, accepting the additional host-management responsibility.

USB radios become part of the server architecture

Zigbee, Thread and Z-Wave coordinators must remain stable across reboots and updates. Bare-metal USB attachment is simple; virtual machines need explicit passthrough. Network-based coordinators can reduce the dependence on a particular hypervisor USB mapping.

Keep radio signal quality in mind. A coordinator physically jammed against the server, USB 3 noise or a metal cabinet can cause problems that no amount of CPU performance will solve.

The Frigate integration is a network relationship

Home Assistant and Frigate do not have to be the same process or even the same machine. Frigate documents local-network integration URLs and same-device Docker patterns. That flexibility lets you split the workloads later without rebuilding the entire smart home.

Use stable hostnames or reserved IP addresses and design the network so Home Assistant can reach Frigate reliably. If the server hosts both, keep the internal Docker or VM network understandable rather than relying on fragile ad-hoc port mappings.

Recording storage should not crowd Home Assistant

Frigate recordings can consume terabytes, while Home Assistant configuration and databases are comparatively small. Put the recording volume on storage designed for that retention job and keep enough free SSD capacity for the operating environment.

A nearly full recording disk should not prevent Home Assistant from writing its database or installing an update. Separate volumes and clear retention rules improve both reliability and troubleshooting.

PoE and camera networking still matter

The server is only one component of the NVR. A PoE switch must have enough ports and total power budget, and camera traffic should use reliable wired links. VLANs can improve camera isolation when the network is managed accordingly.

Cloudzat’s Frigate storage/network Sprint handles those calculations in depth; this combined page focuses on making sure the compute node is not chosen independently of the network that feeds it.

Power protection should keep the automation chain alive

A battery-backed server cannot record cameras if the PoE switch and router lose power immediately. If brief outages matter, put the Home Assistant host, camera switch and core network on a coordinated UPS plan.

This also helps Home Assistant automations remain available during short utility interruptions. Size from actual watts and required runtime rather than the UPS VA number alone.

A good mini PC has serviceable storage and memory

A low price can hide soldered RAM, one tiny SSD and no useful accelerator slot. For a shared Home Assistant/Frigate host, replaceable memory and storage provide a much safer upgrade path.

2.5GbE is a useful extra when recordings or backups move to NAS, while dual Ethernet can help certain network layouts. Neither feature compensates for missing video acceleration or inadequate detector support.

Know when to split Home Assistant and Frigate

One box is convenient, but failure-domain consolidation has a cost. A driver update, storage failure or hypervisor outage can take both automation and surveillance offline.

Split the services when cameras are mission-critical, when Home Assistant must remain exceptionally stable, or when the Frigate GPU/driver stack changes frequently. The best combined server is one that can later become two servers without a painful migration.

Questions people ask

Best Hardware for Home Assistant and Frigate questions

What CPU is good for Home Assistant and Frigate?

A recent Intel N100/N150 can handle a small well-accelerated setup; N305 or Core-class systems provide more shared-service headroom.

How much RAM should the shared server have?

16GB is a sensible baseline. Choose 32GB for more cameras, enrichments or many additional containers.

Can Home Assistant OS and Frigate run on the same machine?

Yes through virtualization or other supported architectures. Home Assistant OS can run in a VM while Frigate runs separately with hardware access.

Can Home Assistant Container run beside Frigate?

Yes on a Linux/Docker host. You manage the host and updates yourself, and Home Assistant Container does not provide the Home Assistant apps system.

Do I need Coral?

No. Frigate supports newer detector paths such as Hailo, Intel OpenVINO and GPUs. Coral remains useful in some installations.

What matters more, camera count or activity?

Both, but simultaneous activity strongly influences detector demand. Several busy cameras can be harder than more mostly quiet cameras.

Should I pass a Zigbee stick into the Home Assistant VM?

That is a common design. Make the USB mapping stable, or consider a network coordinator when you want less dependence on USB passthrough.

Can Frigate recordings share the Home Assistant SSD?

They can, but separate bulk recording storage is easier to size and prevents retention from crowding the Home Assistant environment.

Does Home Assistant need 2.5GbE?

No. The combined server may benefit from it for Frigate playback, NAS recording and backups.

When should I use two machines instead?

Split them when reliability, heavy camera load, GPU driver changes or maintenance windows make a single failure domain unacceptable.

Official references and methodology

Verify current Home Assistant, Frigate and Ollama support

Cloudzat treats Home Assistant, Frigate and local AI as separate workloads that share one hardware envelope. Home Assistant installation guidance, Frigate device-access and detector guidance, and Ollama GPU/context documentation are used as the technical baseline. Calculator outputs are planning tiers rather than benchmark guarantees, and live Amazon listings come only from this sprint's dedicated catalogue.

As an Amazon Associate, Cloudzat may earn from qualifying purchases. Prices, seller terms, exact configurations and Home Assistant, Frigate or Ollama hardware support can change.