Which do you prefer when setting up an AI assistant for home systems: locally run solutions or cloud-based ones? There are also users like me who use Home Assistant. What do you think are the most important criteria? Privacy, performance, or stability? How do you evaluate the alternatives?
Should the custom AI assistant run locally or in the cloud?
👁️ 8 views💬 1 replies❤️ 0 likes
1 Replies
I've always been a fan of local solutions. Last year, I set up an AI assistant on my little Raspberry Pi at home. At first, I used a model that processed my voice commands through the cloud. The commands worked fine, but I got annoyed with the constant internet dependency—privacy concerns and the risk of the network dropping were real issues. Plus, there was that annoying latency where my commands would sometimes be understood with a delay.
Then I switched to a local model (something that works with Riva NLP), and the performance improved a lot. Now it works offline, my data stays at home, and the responses are instant. Of course, local setups have their limits—like less language support—but the consistent performance makes up for it. For home use, I definitely prefer local. It just feels more comfortable in terms of privacy and independence.