News for hackers. Stuff that matters.

A comment on 'Mind-blowing' IBM chip speeds up AI. Parent comment

Re:'Mind-blowing' IBM chip speeds up AI by loufe on Friday October 20, 2023 @01:17PM (#37958836)
That's not true, there are some very pruned (and relatively dumb) LLMs which go under a gig.
[ Parent | on HN ]
Re:'Mind-blowing' IBM chip speeds up AI by andy_ppp on Friday October 20, 2023 @01:24PM (#37958942)
What’s the point of a state of the art AI chip that can’t run large models? It seems problematic to say the least!
[ Parent | on HN ]
Re:'Mind-blowing' IBM chip speeds up AI by nomel on Friday October 20, 2023 @01:39PM (#37959140)
Realtime, low power, isn't really possible today. Think anything motion or reaction related.
[ Parent | on HN ]
Re:'Mind-blowing' IBM chip speeds up AI by superjan on Friday October 20, 2023 @01:32PM (#37959054)
[delayed]
[ Parent | on HN ]
Re:'Mind-blowing' IBM chip speeds up AI by Jensson on Friday October 20, 2023 @01:30PM (#37959011)
It can run small models very fast.
[ Parent | on HN ]
Re:'Mind-blowing' IBM chip speeds up AI by jacobgorm on Friday October 20, 2023 @03:14PM (#37960244)
Some of us still work on computer vision. 224MiB is fairly massive for convolutional neural network.
[ Parent | on HN ]
Re:'Mind-blowing' IBM chip speeds up AI by rubatuga on Friday October 20, 2023 @03:18PM (#37960288)
Small LLMs or large small language models???
[ Parent | on HN ]