LocalLLaMA

3 readers

1 users here now

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

founded 1 year ago

MODERATORS

communick@poweruser.forum

Comparing 4060 Ti 16GB + DDR5 6000 vs 3090 24GB: looking for 34B model benchmarks (alien.top)

submitted 1 year ago by regunakyle@alien.top to c/localllama@poweruser.forum

8 comments fedilink hide all child comments

I am going to build a LLM server very soon, targeting 34B models (specifically phind-codellama-34b-v2.Q4 GGUF GPTQ AWQ).

I am stuck between these two setups:

12400 + DDR5 6000MHz 30CL + 4060 Ti 16GB (GGUF; Split the workload between CPU and GPU)
3090 (GPTQ/AWQ model fully loaded in GPU)

Not sure if the speed bump of 3090 is worth the hefty price increase. Does anyone have benchmarks/data comparing these two setups?

BTW: Alder Lake CPUs run DDR5 in gear 2 (while AM4 run DDR5 in gear 1). AFAIK gear 1 offers lower latency. Would this give AM4 big advantage when it comes to LLM?

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

[–] mcmoose1900@alien.top 1 points 1 year ago (1 children)

Here's a 7B llama.cpp bench on a 3090 and 7800X3D, with CL28 DDR5 6000 RAM.

All layers offloaded to GPU:

Generation:5.94s (11.6ms/T), Total:5.95s (86.05T/s)

And here is just 2/35 layers offlloaded to CPU:

Generation:7.59s (14.8ms/T), Total:7.75s (66.10T/s)

As you can see, the moment you offload even a little bit to CPU, you are going to hit performance hard. More than a few layers and the hit is very severe.

Here is exllamav2 for reference, though the time also includes prompt processing so its actually faster than indicated:

3.91 seconds, 512 tokens, 130.83 tokens/second (includes prompt eval.)

[–] regunakyle@alien.top 1 points 1 year ago

Thanks for your data! Can you do the test again with the phind codellama 34B model?