Microsoft Surface Laptop Ultra for Business Review: RTX Spark N1X Runs a 35B-Parameter Coding Agent at 89 Tokens per Second, With One Major Caveat
The Microsoft Surface Laptop Ultra for Business is the first Surface laptop built on NVIDIA silicon, and the first RTX Spark system to get a review from us. Our unit carries the full chip: a 20-core Grace Arm CPU, a 6,144-core Blackwell RTX GPU, and 128GB of LPDDR5X unified memory, with the user able to The post Microsoft Surface Laptop Ultra for Business Review: RTX Spark N1X Runs a 35B-Parame…
A 35B-parameter coding agent running at 89 tokens per second on a laptop with 128GB of LPDDR5X unified memory shows how much local AI inference now depends on memory capacity and bandwidth, not just compute. The Surface Laptop Ultra for Business pairs a 20-core Grace Arm CPU with a 6,144-core Blackwell GPU on one shared memory pool, a design that pushes workloads toward the edge that were once reserved for server-class hardware. For enterprise buyers, that could shift how teams split inference between endpoints and datacenter nodes, and it keeps pressure on high-capacity DDR5, HBM, and MRDIMM supply for the workloads that still need to scale up. The review also flags one major caveat, so it's worth weighing before assuming endpoint hardware can replace server capacity. If you're planning memory for AI infrastructure, check live inventory or request a quote to see what's available now.