The short version
● Form factor — PCIe FH3/4L at 150W TDP
● Compute — 870 TOPS INT8
● Memory — 128GB LPDDR4x at 548GB/s
● On-die SRAM — 576MB
Qualcomm's product brief for the Cloud AI 100 Ultra puts it in a PCIe FH3/4L form factor at a 150W TDP, rated at 870 TOPS INT8, with 128GB of on-card LPDDR4x at 548GB/s and 576MB of on-die SRAM. Qualcomm describes the card as fully programmable with support for recent AI techniques and data formats, and does not publish a supported precision list on that brief beyond the INT8 rating.
The number that matters for inference work is the 128GB at 150W. That capacity in a single-slot, 150W envelope is aimed squarely at serving large models where memory capacity, not raw compute, is the binding constraint, and where rack power budgets decide how many cards fit per node. It is a different proposition from a high-TDP training accelerator, and it is the figure worth comparing when you are sizing an on-premises inference box rather than a training cluster.
More from 5 November
Full dated index: every AI release in November 2026.
Sources