I get a diff and shit works

Every local model I’ve run has been garbage code and troubleshooting. I only use local for non technical stuff. Using qwen2b with your local configuration seems primed for hassle.

I have Hermes installed and use Claude for all troubleshooting. Works absolutely fantastic telling me what’s wrong.

It’s just files, how hard could it be 🤷

Humans definitely don’t make mistakes. Especially not with their jank home servers. All of it done without shortcuts and auditable in code, with a standard process.

You’ve set some imaginary high bar that is rarely actually met outside of nerds shit talking on the internet and a small percentage of users.

I bet you I do less work and have a more efficient and locked down network, file permissions, fully rootless, updated on the hour, 100% in code, and observable setup than 98% of you just running sudo apt upgrade and praying. But yeah, my little agent working in an nemoclaw sandbox is going to go rogue and destroy my server somehow and I definitely believe the AI hype bros that are definitely “accidentally “ are hacking orgs and not going to end the human race🙄

Man, the ignorance and prejudice over how llm work is insane.

Hermes + maybe $5-10 worth of tokens if I need more compute than ollama can provide (on a busy month). Running everything in a nice podlet setup and have reduced maintenance down to trivial levels. Config gets occasionally shipped to GitHub

In theory I might lose the ui if it bricked itself but in reality that doesn’t happen.

Removed by mod

Removed by mod

Removed by mod

Removed by mod

Removed by mod

Has been for years now at this point. Try harder to justify anti ai stance 🤣

Bro finds a Reddit thread and instant story, lemmy nerds pick it up and actually believe more than 5 people noticed or gave a fuck about an update 🙄

AI slop coming for ya, amirite folks?

Removed by mod

Removed by mod

Removed by mod

The vegans and crossfitters of the tech world.

So made up fears over settled law. Makes perfect sense.