NEWThe Builders section →
Run local AI.
Free toolOpen →
Can your PC run it?
RTX 3060 · 12 GB · 32 GB RAM
Qwen 3 8B~35 t/s
Qwen 3 14B~20 t/s
Gemma 3 27Boffload
Latest build
Colibri Runs a 744B AI Model on 25 GB of RAM. What It Means for Your Homelab
Read the build →
My homelab
UniFi Gateway UltraIntel NUC 13 ProMac Mini M4Synology DS723+
See the rig →
Run local AI.
Self-host
everything.
Real builds, step-by-step guides and free tools for running AI and your own apps on hardware you own.
Builders
All Builders articles →
Build with AI, ship at home.
Builders · 4 min
7 Rules for Letting an AI Agent Maintain Your HomelabI let an AI coding agent edit my live server and clean up my NAS. These are the guardrails that make it…
27 Sep · 4 min
How I Run Claude Code 24/7 on a Mac Mini With tmux
27 Sep · 4 min
12 GitHub URL Tricks Every Homelabber Should Know
Fresh from the lab
All guides →
docker & infrastructure
Docker & Infrastructure · 26 SepHow to Integrate Portainer with Prometheus for Container Metrics
ai media & transcription
AI Media & Transcription · 24 SepHow to Install Whisper.cpp on Ubuntu: Self-Hosted Setup Guide
ai media & transcription
AI Media & Transcription · 23 SepBazarr Docker Compose Setup: My Homelab Config with Explanations
docker & infrastructure
Docker & Infrastructure · 22 SepHow to Integrate CasaOS with Grafana for System Monitoring
ai coding tools
AI Coding Tools · 21 SepTabby Docker Compose Setup: My Homelab Config with Explanations