@zkatkat Looking forward to reading the "Why not LLMs?" list since these instances seem to vary a lot in degrees of concerning. For example, the Calibre changlog makes it sound like it's simply adding functionality to call local models via LM Studio. That's way different than projects that use LLM-generated code. I think? Genuinely looking forward to reading more.
If you have a fediverse account, you can quote this note from your own instance. Search https://mas.to/users/basetwojesus/statuses/115861304510447580 on your instance and quote it. (Note that quoting is not supported in Mastodon.)