[NVIDIA Adjusts Rubin GPU Specifications: HBM and NVLink Bandwidth Reduced, Network Bandwidth Increased] According to a report by DeepFlow TechFlow on October 26th, comparing NVIDIA's October 2026 documentation with a third-party-saved June document reveals that NVIDIA has reduced the HBM memory bandwidth and NVLink bandwidth of its Rubin GPU systems, while increasing network bandwidth. Specifically, the HBM4 memory bandwidth of the Vera Rubin superchip has been reduced from 44 TB/s to 38.5 TB/s, and the NVLink bandwidth from 7.2 TB/s to 6 TB/s. Meanwhile, the scale-out network bandwidth has been increased from 0.8 TB/s to 0.9 TB/s. The HBM capacity of the HGX NVL8 system has been reduced from 2.3 TB to 2 TB, and the bandwidth has decreased from 176 TB/s to 130 TB/s. The NVLink switch bandwidth has also been reduced from 28.8 TB/s to 24 TB/s. The report points out that lowering the HBM4 memory bandwidth threshold is expected to allow more memory products to meet the specifications of Rubin GPUs, thereby increasing final production volume. Furthermore, NVIDIA has adjusted the memory specifications for Vera CPUs to support a maximum of 1.5 TB of LPDDR5X per chip.