new | show | ask | jobs Github

tapvt 10 hours ago

LLMs are stateless. But given an actual database of task-shaped items and some work, I could see the potential.

With a canonical source of truth, and set input/output expectations, the potential blast radius is quite small.

▲

wpm 9 hours ago | parent [-]

And the end results is.....? What? A todo app that takes 16GB of RAM?

▲

bigyabai 8 hours ago | parent [-]

Nothing that Mac and Windows users aren't already used to.

▲

Forgeties79 7 hours ago | parent [-]

It’s tempting to be flippant about MacOS/windows but in all seriousness, the resources required for an LLM to do the job of a typical lighter weight app/software is a serious consideration. No amount of bloat matches what an LLM needs.

▲

bigyabai 5 hours ago | parent [-]

> No amount of bloat matches what an LLM needs.

I don't think that's necessarily true. For instance, LinkedIn uses more memory than Gemma E2B inference does.

▲

Forgeties79 4 hours ago | parent [-]

LinkedIn is an entirely different category and an extreme case at that. We’re not talking about LLM’s replacing LinkedIn either. It’s an entirely different comparison/discussion.

	▲	cyanydeez 2 hours ago \| parent [-]
		Finally, we've fully documented the Singularity-is-actually-just-bloated software.