/local-ai/home-automation
AI Energy Management With Home Assistant and EMHASS
Use EMHASS with solar forecasts, dynamic tariffs, EV charging constraints and deterministic Home Assistant fallbacks.
< Back to all guides
Topic archive
Local AI
Self-hosted models, security AI, private inference and practical hardware benchmarks.
/local-ai/home-automation
Build an Offline Multimodal Voice Assistant for Home Assistant
Build a local Home Assistant voice pipeline with wake word, speech, Gemma 4 context, allowlisted actions and spoken responses.
/local-ai/local-first
Gemma 4 Audio Local Inference: What Works Offline and What Does Not
Test Gemma 4 audio locally for transcription and translation, with realistic runtime, format, memory and latency limits.
/local-ai/local-first
Local ChatGPT: model size in GB and 16–32 GB RAM builds
How many GB does local AI need? Compare 2.0 GB and 4.9 GB model downloads, RAM and VRAM needs, and what differs from using ChatGPT online.
/local-ai/local-first
Local RAG With Open WebUI: Chat With Documents Without the Cloud
Build an offline Open WebUI RAG workflow for PDFs with local embeddings, practical chunking, evidence checks and private storage.
/local-ai/local-first
Open WebUI on Windows: Build a Private ChatGPT With Ollama
Run Open WebUI and Ollama on Windows with persistent storage, LAN access controls, backups and no accidental internet exposure.
/local-ai/home-automation
Reduce Food Waste With Local OCR, Private RAG and Fridge Photos
Turn fridge photos into a private local inventory with OCR, recipe retrieval and honest date-label alerts that require confirmation.
/local-ai/home-automation
Build a Local AI Irrigation System With Soil Sensors and Weather RAG
Combine calibrated soil sensors, local climate history and guarded Home Assistant automations for private, water-aware irrigation.
/local-ai/home-automation
Build an Offline Care Assistant for Older Adults or Pets
Combine local voice, fall signals, medication reminders and family alerts without turning a home lab into a medical device.
/local-ai/local-first
Qwen3 vs Gemma 4 for Local Coding: Which Model Fits Your Hardware?
Compare Qwen3 and Gemma 4 for private C# and .NET coding by memory use, context, tooling and reproducible local tests.
/local-ai/local-first
How to Run a Local Coding LLM With Only 4GB or 8GB of VRAM
Run a useful local coding model on limited VRAM by controlling quantization, context, KV cache and CPU offload.
/local-ai/home-automation
Best GPU for Home Assistant AI in 2026: VRAM, Power and Real Costs
Compare RTX 3050 6/8 GB, RTX 3060 12 GB, RTX 4060 and RTX 5070 for local LLMs, Whisper and camera AI, including power and case-fit trade-offs.
/local-ai/home-automation
Offline Voice Control for Home Assistant With Gemma 4
Design a private voice control stack for Home Assistant using local speech, Piper output, and a constrained Gemma 4 action layer.
/local-ai/local-first-storage
Immich Hardware Requirements: Mini PC, NAS, or Old Desktop?
Choose a mini PC for Immich: official RAM requirements, a practical 16 GB setup, local SSD vs NAS storage, and separate backups.
/local-ai/home-automation
Why Local Home Automation Gets Better With a Local AI Layer
Use Home Assistant, Zigbee, and a local Gemma 4-style model to make home automation more private, resilient, and useful.
/local-ai/local-first-storage
Build a Local NVR With Frigate and No Cloud Subscription
Learn how a local-first NVR with Frigate can record security cameras at home without relying on paid cloud storage.