I have dabbled with running llms locally, I’d absoluteley love to but for some reason amd dropped support for my GPU in their ROCm drivers, which are needed for using my GPU for ai on Linux.

When I tried it fell back to using my cpu and I could only use small models because of the low vram of my RX 590 😔

By the way…

Holy comment section batman

This is the only good answer. No need to distract yourself by figuring out your speed and guesstimating your gap like others are saying. Just count the seconds whenever you need to

Please visit your local democracy officer for re-education

Option 2 would be your best bet. Great balance between security and convenience. Bitwarden is my go to because afaik it stores every detail encrypted (unlike mainstream PWs) and when you open your vault, the database gets transferred to your pc and is decrypted locally. Its essentially the same as option 1, just 1000x more convenient.

Id only selfhost vaultwarden if you want bitwardens premium features, if you dont then youre maintaining a service which you wouldnt really need. Not to mention if you selfhost on a machine on your network, you have to deal with exposing that machine to the internet, not really worth it imo.

I love this comment section. Perfect balance of people actually trying to discuss and people getting pissy because their answer can be the only right one.

I don’t know shit but when I try and picture it happening in my head I can only see A being correct, think Doctor Strange portals. When he moves a portal the person comes out stationary