logoalt Hacker News

rynnyesterday at 11:28 PM1 replyview on HN

> Please do give that a try and report back the prefill and decode speed.

M4 Max here w/ 128GB RAM. Can confirm this is the bottleneck.

https://pastebin.com/2wJvWDEH

I weighed about a DGX Spark but thought the M4 would be competitive with equal RAM. Not so much.


Replies

cmrdporcupineyesterday at 11:33 PM

I think the DGX Spark will likely underperform the M4 from what I've read.

However it will be better for training / fine tuning, etc. type workflows.

show 1 reply