NVIDIA HGX Platform

Supercharging AI and high-performance computing for every data center.

Explore Systems

Overview

Supercharging AI and HPC for Every Data Center

The NVIDIA HGX™ platform brings together the full power of NVIDIA GPUs, NVIDIA NVLink™, NVIDIA networking, and fully optimized AI and high-performance computing (HPC) software stacks to provide the highest application performance and drive the fastest time to insights for every data center.

The NVIDIA HGX Rubin NVL8 integrates eight NVIDIA Rubin GPUs with sixth-generation high-speed NVLink interconnects, delivering 5.5x more NVFP4 FLOPS than HGX B200 to propel the data center into a new era of accelerated computing and generative AI.

NVIDIA Kicks Off the Next Generation of AI With Rubin — Six New Chips, One Incredible AI Supercomputer

The next-generation platform scales mainstream adoption, slashing cost per token with five breakthroughs for reasoning and agentic AI models.

Read the Press Release

AI Reasoning Performance and Versatility

AI, complex simulations, and massive datasets require multiple GPUs with extremely fast interconnections and a fully accelerated software stack. The NVIDIA HGX™ platform brings together the full power of NVIDIA GPUs, NVIDIA NVLink™, NVIDIA networking, and fully optimized AI and high-performance computing (HPC) software stacks to provide the highest application performance and drive the fastest time to insights for every data center.

Unmatched End-to-End Accelerated Computing Platform

The NVIDIA HGX B300 integrates NVIDIA Blackwell Ultra GPUs with high-speed interconnects to propel the data center into a new era of accelerated computing and generative AI. As a premier accelerated scale-up platform with up to 11x more inference performance than the previous generation, NVIDIA Blackwell-based HGX systems are designed for the most demanding generative AI, data analytics, and HPC workloads.

NVIDIA HGX includes advanced networking options—at speeds up to 800 gigabits per second (Gb/s)—using NVIDIA Quantum-X800 InfiniBand and Spectrum™-X Ethernet for the highest AI performance. HGX also includes NVIDIA BlueField®-3 data processing units (DPUs) to enable cloud networking, composable storage, zero-trust security, and GPU compute elasticity in hyperscale AI clouds. 

AI Reasoning Performance and Versatility

Projected Performance subject to change. Token-to-token Latency (TTL) = 12ms, first token latency (FTL) = 2s, input sequence length =32,768, output sequence length = 8192, eight-way HGX B200 vs eight-way HGX Rubin NVL8, per GPU performance, served using disaggregated inference.

Real-Time Agentic AI Inference

HGX Rubin NVL8 delivers up to 3.5× higher inference performance compared to the previous NVIDIA Blackwell generation, powering advanced reasoning models like DeepSeek-R1. With Rubin’s Transformer Engine coupled with sixth-generation NVLink, Rubin accelerates NVFP4 inference to meet the demands of reasoning-based Agentic AI workloads 

Scalable Training for Large AI Models

Next-Level Training Performance

The HGX B300 platform delivers up to 2.6x higher training performance for large language models such as DeepSeek-R1. With over 2 TB of high-speed memory and 14.4 TB/s of NVLink Switch bandwidth, it enables massive-scale model training and high-throughput inter-GPU communication.

Projected performance subject to change. Perf per GPU, FP8, 16K BS, 16K sequence length.

Accelerating HGX With NVIDIA Networking

AI factories and supercomputing centers span thousands of GPUs as a single distributed computing engine. To keep accelerators fully utilized, AI and scientific workloads demand deterministic latency, lossless throughput, stable iteration times, and the ability to scale not only within a data center but also across multiple sites.

NVIDIA networking delivers the full-stack fabric that makes this possible, combining NVIDIA NVLink scale-up, NVIDIA Quantum InfiniBand and Spectrum-X™ Ethernet scale-out, Spectrum-XGS Ethernet multi-data-center scale-across, NVIDIA® BlueField® DPU and DOCA™ for infrastructure services, and next-generation silicon-photonics platforms—enabling the world’s most demanding AI data centers.

NVIDIA HGX Specifications

NVIDIA HGX is available in a single baseboard with eight NVIDIA Rubin, NVIDIA Blackwell, or NVIDIA Blackwell Ultra SXMs. These powerful combinations of hardware and software lay the foundation for unprecedented AI supercomputing performance.

HGX Rubin NVL8*
Form Factor 8x NVIDIA Rubin SXM
NVFP4 Inference 400 PFLOPS
NVFP4 Training 280 PFLOPS
FP8/FP6 Training 140 PF
INT8 Tensor Core<sup>1</sup> 2 PFLOPS
FP16/BF16 Tensor Core<sup>1</sup> 32 PFLOPS
TF32 Tensor Core<sup>1</sup> 16 PFLOPS
FP32 1040 TFLOPS
FP64/FP64 Tensor Core 264 TFLOPS
FP32 SGEMM | FP64 DGEMMCore<sup>2</sup> 3200 TF | 1600 TF
Total Memory 2.3 TB
NVIDIA NVLink Sixth generation
NVIDIA NVLink Switch NVLink 6 Switch
NVLink GPU-to-GPU Bandwidth 3.6 TB/s
Total NVLink Switch Bandwidth 28.8 TB/s
Networking Bandwidth 1.6 TB/s

* Preliminary specification, subject to change
1. Specification in Dense.
2. Peak performance using tensor core-based emulation algorithms.

HGX Rubin NVL8
Form Factor 8x NVIDIA Rubin SXM
FP4 Tensor Core<sup>1</sup> 400 PFLOPS | 144 PFLOPS
FP8/FP6 Tensor Core<sup>2</sup> 272 PFLOPS
INT8 Tensor Core<sup>2</sup> 4 PFLOPS
FP16/BF16 Tensor Core<sup>2</sup> 64 PFLOPS
TF32 Tensor Core<sup>2</sup> 32 PFLOPS
FP32 1040 TFLOPS
FP64/FP64 Tensor Core 264 TFLOPS
Total Memory 2.3 TB
NVIDIA NVLink Sixth generation
NVIDIA NVLink Switch™ NVLink 6 Switch
NVLink GPU-to-GPU Bandwidth 3.6 TB/s
Total NVLink Bandwidth 28.8 TB/s
Networking Bandwidth 1.6 TB/s
Attention Performance<sup>3</sup> <awaiting results>2x
HGX B300 HGX B200
Form Factor 8x NVIDIA Blackwell Ultra SXM 8x NVIDIA Blackwell SXM
FP4 Tensor Core<sup>1</sup> 144 PFLOPS | 108 PFLOPS 144 PFLOPS | 72 PFLOPS
FP8/FP6 Tensor Core<sup>2</sup> 72 PFLOPS 72 PFLOPS
INT8 Tensor Core<sup>2</sup> 3 POPS 72 POPS
FP16/BF16 Tensor Core<sup>2</sup> 36 PFLOPS 36 PFLOPS
TF32 Tensor Core<sup>2</sup> 18 PFLOPS 18 PFLOPS
FP32 600 TFLOPS 600 TFLOPS
FP64/FP64 Tensor Core 10 TFLOPS 296 TFLOPS
Total Memory 2.1 TB 1.4 TB
NVIDIA NVLink Fifth generation Fifth generation
NVIDIA NVLink Switch™ NVLink 5 Switch NVLink 5 Switch
NVLink GPU-to-GPU Bandwidth 1.8 TB/s 1.8 TB/s
Total NVLink Bandwidth 14.4 TB/s 14.4 TB/s
Networking Bandwidth 1.6 TB/s 0.8 TB/s
Attention Performance<sup>3</sup> 2x 1x

1. Specification in Sparse | Dense
2. Specification in Sparse. Dense is ½ sparse spec shown.
3. vs. NVIDIA Blackwell.

HGX B300 HGX B200
Form Factor 8x NVIDIA Blackwell Ultra SXM 8x NVIDIA Blackwell SXM
FP4 Tensor Core<sup>1</sup> 144 PFLOPS | 108 PFLOPS 144 PFLOPS | 72 PFLOPS
FP8/FP6 Tensor Core<sup>2</sup> 72 PFLOPS 72 PFLOPS
INT8 Tensor Core<sup>2</sup> 3 POPS 72 POPS
FP16/BF16 Tensor Core<sup>2</sup> 36 PFLOPS 36 PFLOPS
TF32 Tensor Core<sup>2</sup> 18 PFLOPS 18 PFLOPS
FP32 600 TFLOPS 600 TFLOPS
FP64/FP64 Tensor Core 10 TFLOPS 296 TFLOPS
Total Memory 2.1 TB 1.4 TB
NVIDIA NVLink Fifth generation Fifth generation
NVIDIA NVLink Switch™ NVLink 5 Switch NVLink 5 Switch
NVLink GPU-to-GPU Bandwidth 1.8 TB/s 1.8 TB/s
Total NVLink Bandwidth 14.4 TB/s 14.4 TB/s
Networking Bandwidth 1.6 TB/s 0.8 TB/s
Attention Performance<sup>3</sup> 2x 1x
Read the NVIDIA Blackwell Ultra Datasheet
Read the NVIDIA Blackwell Datasheet

Learn more about the NVIDIA Rubin Platform.

Learn More
Products
Technologies
Resources
Company Info