Turn your Mac into the brain of your household and operate from your phone.
Prajna — on-device memory-augmented reasoning: a 6.7M Cognitive Resonance Network inside frozen Gemma 4 E2B. Passes CEHRI 60/60 via retrieval memory, 91.7% on unseen reworded exams. Trained on a Mac Mini M4, no GPU, no API.
GitHub Actions cron that watches Apple's refurb store for an M4 Mac mini under a price cap and pings Slack.
Open a browser tab — you're now driving your computer. Live remote desktop + real PTY shell (Claude Code, vim, top) + on-device voice. macOS and Windows, every machine you own behind one URL and one password. Self-hosted.
NotchX is adds MacBook like notch to Windows ,Linux and Mac Desktop devices at the same time providing better user experience
Memory-safe, LAN-accessible OpenAI-compatible server for Gemma 4 12B (4-bit MLX) on a 16 GB M2 Pro Mac mini
Blind, task-oriented eval: can a local LLM actually WRITE as well as a frontier model? Measures output quality, not tokens/sec.
A platform-agnostic format + method for sizing local-LLM hardware from your real agent sessions, calibrated with capability-oracle runs.
Give your local AI eyes and memory — native to your Mac. Apple Silicon LLM (MLX) + web proxy + conversation RAG in one process, OpenAI-compatible.
Free, self-hosted, ultra-low-latency Push-to-Talk voice typing & AI translation for Windows, offloaded to an Apple Silicon Mac via Apple MLX & Ollama.
MacBook + Mac mini as one system: your laptop is the console, the mini is the always-on runtime. SSH wiring, cron that actually fires, and a watchdog that makes silence trustworthy.
Headless Mac mini fan control bridge: spend cheap fan life to protect expensive storage thermal conditions.
🤖 Personal AI stack on Mac: Ollama + Open WebUI + Stable Diffusion + TripoSR 3D + Whisper STT + Tax AI Social + Gmail AI + Quant AI — 100% local, no cloud
Turn a shelf of Mac Minis into a self-coordinating Claude Code worker fleet. Each job opens a real visible terminal you can watch with your skills preloaded; the database is the only coordination layer; workers claim, heartbeat, and reap each other through Postgres advisory locks. Scale from one machine to many with a single skill invocation.
Karpathy's autoresearch running on Mac Mini — no NVIDIA GPU required. Auto-detects CUDA, Apple Silicon (MPS), or CPU.
Three-tier Gemma 4 local inference on Mac Mini M4 — MLX + llama.cpp + Tailscale
Local-first assistant for safely capturing, classifying, and organizing Telegram/OpenClaw inputs in an Obsidian vault.
Self-hosted AI platform on Apple Silicon — production-proven, Thai-optimized, $3.50/day
Local-first AI that turns your digital life into timely, evidence-backed actions. Nothing fake.
To associate your repository with the mac-mini topic, visit your repo's landing page and select "manage topics."