LocalLLaMA

1 readers

1 users here now

Community to discuss about Llama, the family of large language models created by Meta AI.

founded 10 months ago

MODERATORS

communick@poweruser.forum

AMD EPYC CPU or 1x RTX 4090? (alien.top)

submitted 10 months ago by M000lie@alien.top to c/localllama@poweruser.forum

13 comments fedilink hide all child comments

I plan to infer 33B models at full precision, 70B is second priority but a nice touch. Would I be better off getting an AMD EPYC server cpu like this or a RTX 4090? With the EPYC, i am able to get 384GB DDR4 RAM for ~400USD on ebay, the 4090 only has 24GB. Moreover, both the 4090 and EPYC setup + ram cost about the same. which would be a better buy?

you are viewing a single comment's thread
view the rest of the comments

[–] Silent-Edge1414@alien.top 1 points 10 months ago (1 children)

What is the frequency of your ram? DDR4 2666 mhz, DDR4 3200 mhz or DDR5 4800mhz? and how are they installed on the motherboard? 4x128gb(quad channel) or 8x64gb(octa channel)?

Rams frequencies are the most important for llm token generation, as these are often the bottleneck. With a 32 or more-core Epyc 7003 cpu in octa channel (DDR4 3200), you can expect 3 to 4 tokens(70b) equivalent to a 200GB/S vram of speed.

For OP, a 48-core cpu (genoa) or more, in 12-channel DDR5 4800 can expect to go up to 6 to 8 tokens(70b) equivalent to 400GB/S vram of speed.

an rtx 4090 is around 1000GB/s but with only 24gb vram, gpus are generally much faster for prompt processing than PC cpu (over 100 times faster), but I don't know about modern server cpu(Genoa), normally they are faster in prompt processing than PC cpu, as they natively support fp16/BF16 operation.

But take it with a pinch of salt, as I don't have these configurations at hand, so you'll have to ask someone who does.

[–] XTJ7@alien.top 1 points 10 months ago (1 children)

In my case it's an Epyc 7642 with 8x64GB DDR4 2666, so that may be why my generation is significantly slower.

I find anything below 5 tokens per second not really usable, so that's why I stick with my M1 Ultra. It has plenty of really fast RAM and that again explains most likely why it performs so well, if LLMs are that dependend on fast memory.

I also have a 3090 in another machine but that's also just 24gb and I don't want to shell out more money right now for playing with LLMs, if the M1 Ultra is doing good enough :)

[–] runforpeace2021@alien.top 1 points 10 months ago (1 children)

If you like m1 ultra, wait until m3 ultra … . M3 max already smokes M1 Max by 3x the speed in inference.

So expect m3 ultra to be in the 20t/s range

[–] XTJ7@alien.top 1 points 10 months ago

For sure! But the M1 ultra still holds up really well. I doubt I will replace it for another 3 years at the very least. Currently CPUs are progressing at an impressive rate across the board. Would I like an M3 ultra? Sure, but do I really need it? Sadly no :) The upgrade to an M5 ultra will be insane though.