Articles in this section : Intelligence artificielle.
ASUS's local AI range continues to grow. Among the publicly announced developments, the Ascent GX10 gains a 64 GB variant, while the ExpertCenter Pro ET900N G3 occupies the high end with an NVIDIA GB300 platform. These two form factors address very different needs.
Read the article
The ExpertCenter Pro ET900N G3 operates on a very different scale. With its NVIDIA GB300 Grace Blackwell Ultra platform and 748 GB of coherent memory, this ASUS workstation targets AI projects whose requirements exceed those of a compact machine.
Read the article
ASUS expands the Ascent GX10 with a 64 GB unified-memory version. It retains the NVIDIA GB10 platform but reduces the capacity available for models and working data. Above all, it encourages sizing the purchase around a real requirement.
Read the articleRun models close to your data without dedicating an entire tower to the task: that is the ASUS Ascent GX10's proposition. Behind its mini-PC format are an NVIDIA GB10 platform and 128 GB of unified memory. It targets AI development and inference, with software requirements very different from a consumer PC's.
Read the article
NVIDIA wants to expand the use of its computing tools in semiconductor manufacturing. On 14 October, the company introduced cuPotential, a CUDA-X library for simulating interactions between atoms and molecules, available in early access.
Read the article