memossy

It was interesting Aurora used GPU Max & definitely looking forward to Falcon Shores.

I think Gaudi2 was bad timed & they had to build stack, Gaudi3 is where I think we will see mass adoption given availability, way cheaper price/performance & maturer stack.

There is still weird stuff when using them but they are surprisingly solid.

It's a great company & will do well, plenty of demand & B100s/BH200s etc coming

The Hopper stuff is particulalry interesting

Falcon shores next year will be crazy with 300gb VRAM & new lith

We use v4s, v5es & v5ps. Mostly v5ps, very stable int8 training (versus the horror that is fp8 stability)

I think as we go to enterprise workloads the total cost of ownership becomes important.

NVIDIA is still the best for research given ecosystem but once the models are standardised as with transformers/LLaMA and likely multimodal diffusion transformers it then becomes about scale, availability and cost per flop.

The v5es and v5ps are pretty amazing at running SD, giving code for SD3 now to optimise it on those.

v5es are particularly interesting given the millions that will land and the large pod sizes, particularly well constructed for million token context windows.

"For Stable Diffusion 3, we measured the training throughput for the 2B Multimodal Diffusion Transformer (MMDiT) architecture model. Gaudi 2 trained images 1.5x faster than the H100-80GB, and 3x faster than A100-80GB GPU’s when scaled up to 32 nodes. "

I could but I won't as legal stuff :)

I mean open models yo

If we trained it with videos yes but need more GPUs for that.