Qualcomm's Cloud AI 100 Ultra Packs 128GB of Memory Into a 150W Inference Card

Qualcomm Cloud AI 100 Ultra – 150W AI Accelerator Card with 128GB at 548GB/s

Qualcomm's newest data-center inference card targets generative AI and LLMs while drawing just 150W. The Cloud AI 100 Ultra pairs 128GB of LPDDR4x at 548GB/s with 576MB of on-die SRAM and 64 AI cores, reaching up to 870 INT8 TOPS and 288 FP16 TFLOPS over PCIe Gen4. It ships alongside the Qualcomm AI Inference Suite, which offers OpenAI-compatible APIs, RAG, and Kubernetes or bare-metal deployment.