News for hackers. Stuff that matters.

A comment on GPT-4o's Memory Breakthrough (Needle in a Needlestack). Parent comment

Re:GPT-4o's Memory Breakthrough (Needle in a Needlestack) by nopromisessir on Wednesday May 15, 2024 @10:10AM (#40367193)
Used toys to write a working machine vision project over last 2 days.

Key word: working

The bubble is real on both sides. Models have limitations... However, they are not toys. They are powerful tools. I used 3 different SotA models for that project. The time saved is hard to even measure. It's big.

[ Parent | on HN ]
Re:GPT-4o's Memory Breakthrough (Needle in a Needlestack) by SiempreViernes on Wednesday May 15, 2024 @10:43AM (#40367612)
> The time saved is hard to even measure. It's big.

You are aware that this is an obvious contradiction, right? Big times savings are not hard to measure.

[ Parent | on HN ]
Re:GPT-4o's Memory Breakthrough (Needle in a Needlestack) by nopromisessir on Wednesday May 15, 2024 @11:02AM (#40367871)
Right... With precision...

Furthermore... big mountains are easier to weigh v small individual atoms? I think it's a little more complicated than big is easy to measure...

I care little about the precision... I've got other priorities. It's the same as the time the internet saves me... Big. It's obvious.

I stand by my statement. It's hard to measure...

[ Parent | on HN ]