Web Search with Local AI

TL;DR: Web search in LM Mini lets a local AI chat look up current info while you stay in the thread. SearXNG is free and private, but you run it yourself and point LM Mini at it. Pro Search is built in, no setup, and it needs Pro.
Last post I guessed that all those internet questions were really about this, a local AI that can look something up, so here it is.
A local model only knows what it knew the day it was made. Ask it about last week and it will either shrug, or worse, make something up with a straight face.
You know the routine, I am in the middle of a chat and I need one thing from today, so I leave, open Chrome, find it, copy it, and come back to paste it in, if I still remember why I left. (I do not always remember.) I wanted the chat to go and look for me and then let me keep talking.
There are two ways to give the chat that ability. SearXNG is a search engine you set up yourself, it asks the usual search engines for me and hands back the results, without the ads and without building a profile on me. I give LM Mini its address once, and that is the whole relationship. It is free, and it fits the rest of this app, the model is mine and now so is the search.
Pro Search is for when you would rather not set anything up, it is built into LM Mini, it needs Pro, and it is there the moment you want it. Either way, the one catch is the model, it has to be one that knows how to use tools, or it will never go looking.
I still open a browser when I am actually researching. If I need tabs, and to bounce between pages, and to read something long, that is a browser afternoon.
Next post is image generation, because sometimes the answer is a picture. Until then, try looking something current up from the chat you are already in.
Get LM Mini: lmmini.com