---
title: "Nvidia Quietly Lowers Vera Rubin Specs: HBM4 and NVLink 6 Bandwidth Drops by 12–17%"
description: "Nvidia has lowered HBM4 and NVLink 6 bandwidth in the Vera Rubin platform by 12–17%. Adjustments affected NVL72 and HGX Rubin NVL8 systems, while some pages still display old values."
date: 2026-10-28T17:30:45.127Z
lang: en
url: https://xab.info/en/posts/nvidia-vera-rubin-hbm4-nvlink-specs-revised
tags: [nvidia, vera-rubin, hbm4, nvlink, ai-hardware, gpu, datacenter]
author: "Alexander Buran"
publisher: "XAB.info"
---

# Nvidia Quietly Lowers Vera Rubin Specs: HBM4 and NVLink 6 Bandwidth Drops by 12–17%

![Nvidia CEO Jensen Huang holds up the new Vera Rubin board during a presentation](https://xab.info/media/2026/10/28/nvidia-vera-rubin-hbm4-nvlink-specs-revised/nvidia-vera-rubin-hbm4-nvlink-specs-revised-1.webp)

## Key Points

- Nvidia reduced Rubin GPU HBM4 bandwidth from 22 to 19.2 TB/s and NVLink 6 from 3.6 to 3 TB/s
- NVL72 system metrics dropped from 1,580 to 1,400 TB/s for HBM4; NVL8 dropped from 176 to 130 TB/s
- Training performance for NVFP4 decreased from 280 to 270 Pflops, and for FP8/FP6 from 140 to 130 Pflops
- Some Nvidia web pages still contain old specifications; the company has not explained the discrepancies
- According to Wccftech, changes were made to the site archive between September 1 and 11, 2026

Nvidia has revised the technical specifications of its flagship Vera Rubin AI platform, lowering the stated bandwidth figures for HBM4 memory and the sixth-generation NVLink interface. The adjustments affected individual Rubin GPUs, the Vera Rubin NVL72 rack system, and the HGX Rubin NVL8 platform. The most significant reduction affected the total HBM4 memory bandwidth in the eight-GPU configuration. The changes were introduced on the official product pages in September 2026, but Nvidia did not publish a separate statement explaining the reasons for the revision.

### What Changed at the Individual GPU Level

At the time of the platform's announcement in January 2026, the Vera Rubin NVL72 system specifications stated that each Rubin GPU featured 288 GB of HBM4 memory with a bandwidth of 22 TB/s. After the revision, the memory capacity remained unchanged at 288 GB, but the bandwidth was reduced to 19.2 TB/s, representing a 12.7% drop. At the same time, the NVLink 6 interface bandwidth also decreased: from 3.6 to 3 TB/s per GPU, down 16.7%. These figures determine the baseline performance of each accelerator in conjunction with memory and during data exchange between GPUs within a node.

### System Losses in NVL72 and HGX Rubin NVL8

The Vera Rubin NVL72 rack system, combining 72 Rubin GPUs and 36 Vera CPUs, also lost aggregate performance metrics. While Nvidia previously cited a total HBM4 bandwidth of 1,580 TB/s and an NVLink bandwidth of 260 TB/s for it, these figures are now 1,400 and 216 TB/s, respectively. For the HGX Rubin NVL8 platform, which includes eight Rubin GPUs, the adjustments were even more noticeable: the stated HBM4 memory capacity decreased from 2.3 to 2 TB, and memory bandwidth dropped from 176 to 130 TB/s. NVLink Switch bandwidth also declined from 28.8 to 24 TB/s. Notably, the scale-out networking bandwidth for connecting multiple systems into clusters increased instead.

### Model Training Performance Adjustments

In addition to raw memory and interface characteristics, Nvidia also adjusted neural network model training performance metrics. For the NVFP4 format, the value decreased from 280 to 270 petaflops, and for FP8/FP6 formats, from 140 to 130 petaflops. Although absolute numbers still remain at a level unavailable to competing solutions, a 3–7% decrease in the context of the company's multi-year promises may be perceived by analysts as a signal that targets set during the design phase were not fully achieved in mass production.

### Documentation Discrepancies and Change Chronology

The current HGX Rubin NVL8 product page on Nvidia's website already displays the updated, lowered specifications, but the table for the individual Rubin GPU still shows previous figures: 22 TB/s for HBM4 memory and 3.6 TB/s for the NVLink interface. A separate page dedicated to DGX Rubin NVL8 also continues to claim 2.3 TB of memory and 176 TB/s bandwidth. Nvidia has not clarified whether these discrepancies are due to different hardware configurations or outdated documentation. An analysis of archived specification versions conducted by Wccftech showed that changes for the NVL72 system were made between September 1 and September 11, 2026.

### Market Impact and Nvidia's Position

From a commercial standpoint, such adjustments are unlikely to significantly impact Rubin sales. According to reports, Nvidia has a long queue of enterprise customers awaiting the product release, and even the slightly reduced platform specifications remain well above competitors' offerings. Nevertheless, for the industry as a whole, revising the specifications of a flagship AI accelerator a few months before mass shipments creates a precedent that could affect partner trust and expectations surrounding subsequent architecture generations. For now, Nvidia remains silent regarding the reasons for the revision—whether due to HBM4 memory manufacturing constraints, thermal or electrical characteristics, or a reassessment of targets during final validation.

## Fact-Check & Sources

- [Nvidia не смогла выпустить Vera Rubin, которые обещала: память HBM4 замедлилась, а местами даже ужалась window-new](https://3dnews.ru/1150512/nvidia-ne-smogla-vipustit-vera-rubin-kotorie-obeshchala-pamyat-hbm4-zamedlilas-a-mestami-dage-ugalas) - Detailed breakdown of specification changes with exact figures before and after revision, citing the Wccftech archive (Sept 1–11). All numerical data is consistent across source paragraphs.
- [Samsung первой запустила HBM4 в серию — для ускорителей Nvidia Vera Rubin](https://3dnews.ru/1136770/samsung-pervoy-zapustila-hbm4-v-seriyu-dlya-uskoriteley-nvidia-vera-rubin) - Confirms context: HBM4 for Vera Rubin is manufactured by Samsung, explaining potential supply-dependent performance characteristics.

## FAQ

### By how much did Nvidia reduce HBM4 bandwidth in Vera Rubin?

At the individual GPU level, HBM4 bandwidth was reduced from 22 to 19.2 TB/s (−12.7%). In the HGX Rubin NVL8 system, it dropped from 176 to 130 TB/s, and in NVL72 from 1,580 to 1,400 TB/s.

### Did the HBM4 memory capacity change?

At the individual GPU level, capacity remained unchanged at 288 GB. However, for the HGX Rubin NVL8 platform, the stated total capacity decreased from 2.3 to 2 TB.

### When were the specification changes made?

According to archival analysis by Wccftech, changes for the NVL72 system were recorded between September 1 and September 11, 2026.

### Why do some Nvidia pages still show the old specifications?

Nvidia has not explained the reason for the discrepancies. The individual Rubin GPU page and the DGX Rubin NVL8 page still list previous values (22 TB/s, 2.3 TB, 176 TB/s). The company did not specify whether this is due to different configurations or outdated documentation.

### Will the specification reduction affect Rubin sales?

Analysts estimate the impact will be minimal: Nvidia has a long queue of enterprise clients, and even adjusted figures remain significantly higher than competitor offerings.