Curated hardware and hosting picks for running AI and self-hosted services on your own terms. Every product is tested or researched by our team.
AI Accelerators
Plug-and-play Edge TPU for local AI inference. Perfect for Frigate NVR, object detection, and lightweight ML models.
13 TOPS M.2 AI chip. Drop it into your NAS or mini PC for real-time video analytics and AI workloads.
40 TOPS GPU compute for edge AI. Run Ollama, Stable Diffusion, and small LLMs on dedicated hardware.
GPUs for Local AI
The sweet spot for local AI. 12GB VRAM runs Stable Diffusion, Ollama 13B models, and Whisper comfortably.
16GB VRAM unlocks bigger models — Mixtral, Llama 3 70B quantized, Flux image generation. Best bang/buck for AI.
12GB VRAM with higher compute. Excellent for Stable Diffusion XL, video AI, and LLM inference.
Single Board Computers
The ultimate homelab starter. Run Pi-hole, Home Assistant, lightweight AI, and Docker containers.
Keep your Pi 5 cool under sustained load with the official active cooler and case.
Open-source IP-KVM based on Raspberry Pi. Remote BIOS access, virtual media, and full control of headless servers.
Mini PCs & Servers
Compact Proxmox host. Run Docker, VMs, and lightweight AI workloads with 16GB RAM.
Ryzen 9 7940HS, 32GB RAM, dual NVMe. The ultimate mini PC for Proxmox, Kubernetes, and local LLMs.
Enterprise-grade NUC with vPro, dual NVMe, and Iris Xe graphics. Silent and reliable 24/7.
NAS & Storage
CMR drive built for 24/7 NAS use. Ideal for Synology, TrueNAS, and AI media libraries.
Flagship NVMe for Proxmox VMs, Docker volumes, and AI model storage. Insanely fast and reliable.
Budget NVMe for boot drives and Docker volumes. PCIe 5.0 ready.
Turnkey 2-bay NAS for photos, backups, and self-hosted apps. Runs Docker, Plex, and Immich out of the box.
Memory
Networking
Power & UPS
Cloud GPU & Hosting
Affiliate disclosure: Links on this page may earn us a commission at no extra cost to you. We only recommend products we believe in.