Re:Meta Llama 3
by tmikaeld on Thursday April 18, 2024 @01:33PM
(#40078655)
Hm, how much VRAM would this take to run?
[ Parent
| on HN ]
My guess is around 256GiB but it depends on what level of quantization you are okay with. At full 16bit it will be massive, near 512GiB.
I figure we will see some Q4's that can probably fit on 4 4090s with CPU offloading.
With 400 billion parameters and 8 bits per parameter, wouldn't it be ~400 GB? Plus context size which could be quite large.
he said "Q4" - meaning 4-bit weights.
Ok but at 16-bit it would be 800GB+, right? Not 512.
Re:Meta Llama 3
by reactordev on Thursday April 18, 2024 @06:12PM
(#40081310)
Divide not multiply. If a size is estimated in 8-bit, reducing to 4-bit halves the size (and entropy of each value). Difference between INT_MAX and SHORT_MAX (assuming you have such defs).
I could be wrong too but that’s my understanding. Like float vs half-float.
Re:Meta Llama 3
by mrtranscendence on Thursday April 18, 2024 @02:09PM
(#40079025)
Back of the envelope, maybe 0.75TB? More than you have, probably ...