Silicon
AMD MI455X vs NVIDIA Rubin: can more memory beat NVLink?
The MI455X carries 432 GB of HBM4, versus 288 GB on Rubin. AMD still has to prove that UALoE can turn that capacity advantage into rack-scale performance.
GPUs & Silicon
Silicon: GPUs, memory and interconnects
GPUs, HBM memory, NVLink interconnects and microarchitecture: what really sets the performance and cost of AI compute, from the die to the datacenter.
5 articles
Articles: Silicon
Silicon
DGX Spark vs RTX 5090: capacity or speed, the choice that decides your local LLM
The DGX Spark and the RTX 5090 answer two opposite questions: hosting a very large model, or serving a mid-size one fast. The duel in numbers, measurements and mechanisms included.
Silicon
How much VRAM do you need to run an LLM locally?
How much VRAM does a local LLM need? The ~2 GB per billion parameters rule, the weight of the KV cache, what quantization changes, and what actually fits on your card.
Silicon
RTX Spark vs DGX Spark: two different chips, one 273 GB/s wall
The RTX Spark (N1X) is not a rebadged DGX Spark: it is the Grace Blackwell SoC for Windows on Arm, with 128 GB of unified memory. Why the ~273 GB/s, not the "petaflop", decide what it can actually do.
Silicon
TorchTPU, XLA, JAX: how Google attacks NVIDIA's software lock-in
TorchTPU, PyTorch/XLA, JAX, XLA and vLLM form a stack whose keystone is a compiler. How Google attacks the one asset that keeps developers on NVIDIA: the cost of leaving CUDA.