ASUS has announced it will be shipping the Ascent GX10 with 64GB of memory, joining the existing 128GB variant for more flexibility when it comes to shopping for local AI machines.

ASUS Ascent GX10 64GB

The 64GB version uses the same NVIDIA GB10 Grace Blackwell Superchip as the 128GB model. It pairs a Blackwell GPU with fifth-generation Tensor Cores and FP4 support with a 20-core Grace Arm CPU, connected by NVLink-C2C for up to 1 petaflop of AI performance. The brand says this version is enough for today’s most capable 30 to 35B class models and autonomous agents, pointing to models like Qwen3.8-27B that now deliver performance comparable to frontier models from just months ago.

When one unit isn’t enough, two can be clustered using the built-in ConnectX-7 networking and NVIDIA Sync’s Cluster Assistant, with setup taking minutes and no changes to the software environment. That path also supports NVIDIA’s NVFP4 format, which lets users run compressed versions of larger models such as Qwen3.8-Flash-Next.

The thermal design carries over from the 128GB model, with 7-level fan control, ultrawide fins, five heat pipes, and two 140x80mm fans pulling air through bottom vents for sustained performance over long agent sessions. It ships with NVIDIA DGX OS plus PyTorch, Jupyter, and Ollama, along with NVIDIA OpenShell and NemoClaw for policy-based agent security. Built-in SSH-based remote management lets IT teams provision and monitor fleets alongside existing infrastructure.

ASUS’s Malaysia office has confirmed the availability of the new SKU for the local market, with the actual launch date to be revealed in the near future.

-advertisement-
Facebook
Twitter
LinkedIn
Pinterest

Related Posts

-advertisement-
Subscribe via Email

Enter your email address to subscribe to Tech-Critter and receive notifications of new posts by email.