Konraddo

I know this is not the theme of this post, but I wonder if there’s an LLM that doesn’t hallucinate when asked to summarize information of a group of documents. I tried Gpt4all for simple queries like finding out which documents mentioned a certain phrase. It often gave me filenames that didn’t actually exist. Hallucinating contents is one thing but making up data source is just horrible.

You have to use the ‘OS’ version though, or is it my wrong understanding?

Agreed that there’s no all-in-one solution to play local music and music on Spotify (if I’m a premium user). I vaguely recall there’s a solution to automate playing Spotify music and record it in real-time (since you cannot download music directly) but it seemed too troublesome, so I eventually chose spot-dl4 to download music from YouTube using Spotify playlist, then the folder got imported into Lidarr/Navidrome, then my Symfonium on Android connects to Navidrome to get the songs.

It’s quite a bit of manual work to add songs to a separate playlist if I like something on Spotify then use spot-dl4 to do the download. At least, I successfully keep a copy of my favourite songs on my server.

It’s a headache most of the time so you might consider purchasing a local SIM card for 4/5G connection instead (and share connection via mobile phone) in the future.

Depends on the application really. For example, I don’t need to update Jellyfin and the arrs as soon as the new updates drop. They work just fine and I’m not waiting for any particular fixes.

Between Tube Archivist (TA) and Pinchflat (PF), it seems TA is a better choice (because you want to delete the downloaded videos). TA has a built-in interface to watch and delete the video. But if you are like me, who watches the videos in Jellyfin and don’t plan to delete them afterwards, then PF is a solid archival application.