Filtered by topic: Home Assistant Clear ✕
… I'm under no illusions that it'll be hands-off, but it'll reduce my maintenance to a few line items to handle rather than a substantial drain on my time. Time that I can use to put more things into place. I'm sure you've noticed how bad the search experience has become. …
… And that’s just the LLM side… It’s even compatible with AI models besides LLMs Image generators, TTS models, and STT pipelines, Open WebUI can do it all Most productivity-driven AI applications can only harness LLMs to aid my daily tasks, with a few supporting embedding models at most. …
… While the transcription quality is fairly accurate and the features are sometimes useful, it’s hard to ignore the privacy concerns. If a company is willing to bypass consent and send emails on your behalf, it raises questions about how it handles the data it collects. …
… In fact, I’ve got two different AI-powered pipelines set up in my Home Assistant hub – one that lets me manage my smart devices using my voice, while another bridges the gap between local AI apps and HASS by tossing MCP servers into the mix. …
… The modern local AI smart home stack Ollama plus Qwen is the way to go The current local AI smart home stack is surprisingly approachable compared to what it looked like even two or three years ago. …
… I’ll admit that it took me more than a year and hours of troubleshooting to fully deck out my HASS instance. But I’d happily do it again if I had to, as few things are more satisfying than controlling every aspect of my living space through voice commands and automation rules. …
… Related I revived my old Android phone by turning it into a better voice assistant than any smart speaker I've owned This free Home Assistant setup finally made my old Android phone useful again I've used the unofficial HA-MCP Server repo But there’s even an official MCP integration for Home Assist… …
… But since I wanted a full-on voice assistant, I hooked HASS up to a bunch of AI models I host on my home lab nodes. My current conversation agent is Gemma-4-26B-A4B , which runs on my aged hardware thanks to MoE offloading. …
… But those guardrails went in weeks and months after the public pressure started, not before. Security is not a starting constraint for that project, but a phase the maintainers are seemingly graduating to under duress. …
… The ambitious project of running a language model on your own hardware is now a weekend afternoon project. That has permeated smart home automation so swiftly that you can build a setup that rivals what Google pitches as its premium AI feature. That’s not a cheap shot. …