NVIDIA Blackwell Ultra (GB300) is a Blackwell-family GPU optimized for AI reasoning workloads. Compared with standard Blackwell it offers 1.5x larger HBM3E memory and 1.5x more dense FP4 Tensor Core FLOPS.
Blackwell Ultra is the core of the rack-scale GB300 NVL72 system, which integrates 72 Blackwell Ultra GPUs and 36 Grace CPUs (2,592 Arm Neoverse V2 cores) into a single, fully liquid-cooled platform. The system delivers 1,080 PFLOPS of FP4 (with sparsity), 20 TB of GPU memory at up to 576 TB/s, and 17 TB of LPDDR5X CPU memory (14 TB/s); fifth-generation NVLink provides 130 TB/s of total bandwidth.
Unlike GB200 (optimized for scalable LLM inference), GB300 is purpose-built for the age of AI reasoning.
AI Accelerator · serves as: AI Inference, High-level compute.
Which group NVIDIA Blackwell Ultra belongs to and how it is built
Compute Modules is a subcategory of hardware components that provide processing power for robotic systems. It encompasses onboard computers, single-board computers (SBCs), AI accelerators, embedded processors, GPU/NPU compute modules, and other units responsible for processing sensor data and executing control logic. These modules form the foundation of modern autonomous, humanoid, and perception-capable robots.
An AI Accelerator is a specialized hardware component designed for efficient execution of artificial intelligence computations, particularly neural network inference, computer vision processing, and sensor data analysis. In robotics, AI accelerators are used to run perception models, object recognition, image segmentation, planning, and other tasks that require high computational throughput under constrained power budgets. They may take the form of dedicated NPU, TPU, VPU, or GPU chips, or specialized embedded modules.
