Run this on your personal internet connection, risk not being able to use a search engine.
For personal use, you can run it directly on your machine or access over VPN. Queries to upstream search engines can be forwarded over proxies or VPNs as you see fit. Some work fine over tor and some can go over commercial or DIY tunnels.
Config is in a git repo I give access to if requested. One of the technical users modified it to keep pretty minimal logs. I guess they are trusting me to actually use that config but trust is pretty high in the group so not really an issue.
If you run it locally, and only you use it, then you won't get blocked - a given search engine will see about the same number of requests as if you used it directly.
Add a few house members and you'll still be fine.
(I ran the original searx for a year or two locally - no issues at all).
If you delegate queries to e.g. google or bing at that rate, you'll be ip blocked in a heartbeat.
btw thank you for Marginalia! The spirit of the small web is very important to me.
Like I don't mind automated access to my search engine, I even offer a public API to the effect, that you can in fact hook into SearXNG. What I mind is when one jabroni with a botnet decides their search traffic is more important than everyone else's and grabs all the compute for himself via a sybil attack.
Honestly, I just use Kagi. Though I need to find some way to limit my searches to 300 per month.
although existing searx instances have been run for years and they don't seem to be dropping like flies...