I switched from Cursor to Claude Code. Here’s what surprised me.
After months in Cursor, I moved to Claude Code for my homelab scripts. The transition wasn't smooth, but the tradeoffs are worth examining.
Hands-on guides and free tools for builders running LLMs, Immich, Frigate, Paperless-ngx, and Docker on real homelab hardware. No fluff. No paid placements.
Pick your GPU and RAM — see exactly which local LLMs will run, with real tokens-per-second estimates. Updated for 2026.
Select your hardware → see which models you can actually run, with real performance estimates.
Estimates are based on community-reported benchmarks. Actual performance varies by quantization, context length, and backend (Ollama, llama.cpp, vLLM). Amazon links are affiliate links — we may earn a commission at no extra cost to you.
Local LLMs, self-hosted AI apps, automation tools, and hardware — curated and rated for homelab enthusiasts. Run AI on your own terms.
Hands-on guides for securing your homelab, automating your home, and running self-hosted services. Hardening, networking, backups, and privacy — all practical.
Guides, reviews, and deep dives into the tech that powers your home and lab.
After months in Cursor, I moved to Claude Code for my homelab scripts. The transition wasn't smooth, but the tradeoffs are worth examining.
Headscale replaces Tailscale's control plane with your own. Here's what happens when you connect…
Getting Jellyfin behind Traefik is straightforward until it isn't. Here's the config that actually…
I wired Codeium into my local Ollama setup for offline code completion. Here's what…