NVIDIA HGX B200
The NVIDIA HGX B200™ delivers a new era of accelerating computing and generative AI, integrating NVIDIA Blackwell GPUs with high-speed interconnect to power AI performance at scale. As a premier accelerated scale-up x86 platform with up to 15x faster real-time inference performance, 12x lower cost and, 12x less energy usage as compared to Hopper generation GPUs, the NVIDIA HGX B200 is designed for the most demanding AI, data analytics, and high-performance computing (HPC) workloads.

NVIDIA HGX B200
Starting at
$2.890 / Per hour
Reserve now
Cutting-edge acceleration for the world’s most demanding AI and HPC workloads
Pricing

Starting at $2.890 / hour
with 36-month contract, 730 / hour / month

Key features

Accelerated by the groundbreaking NVIDIA Blackwell architecture and Tensor Core technology with new precisions for accelerated AI model training

Connected by NVIDIA 5th Generation NVIDIA NVLink™ 1.8 TB/s interconnect

Dedicated Compression Engine accelerates data queries for large datasets

Enterprise-ready at any scale and any location

Clusters at any size

Vultr's enterprise-ready infrastructure seamlessly supports any cluster size of NVIDIA Blackwell GPUs. Whether you require a small cluster or a massive deployment, Vultr ensures reliable, high-performance computing to meet your specific needs.

Get in touch

Globally available, locally accessible

Large clusters of NVIDIA Blackwell GPUs are available where you need them, thanks to Vultr's extensive infrastructure. With 32 cloud data center regions across six continents, we guarantee low latency and high availability, enabling your enterprise to achieve optimal performance worldwide.

Learn more

Enterprise-grade compliance and security

Vultr ensures our platform, products, and services meet diverse global compliance, privacy, and security needs, covering areas such as server availability, data protection, and privacy. Our commitment to industry-wide privacy and security frameworks, including ISO and SOC 2+ standards, demonstrates our dedication to protecting our customers' data.

Learn more

Download

Purpose-built for AI training, AI inference, and high-performance computing

AI, complex simulations, and massive datasets require multiple GPUs with extremely fast interconnections and a fully accelerated software stack. The NVIDIA HGX™ AI supercomputing platform brings together the full power of NVIDIA GPUs, 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.

No information is required for download

Built for ever-larger language models

NVIDIA HGX B200 delivers generative AI and high-performance computing (HPC) workloads with game-changing performance and efficiency.

As AI models grow in size and complexity, processing the increasing parameters efficiently requires a powerful new GPU. The NVIDIA HGX B200 accelerates growing AI models with less hardware and energy, and with standout performance.

  • GPT-MoE-1.8T training
  • 3x Faster*
  • GPT-MoE-1.8T inference
  • 15x Faster*
  • HPC and Data Analytics
  • 2x Faster*

    • *As compared to NVIDIA Hopper GPUs

Specifications  NVIDIA HGX B2001
Blackwell GPUs 8
Fast Memory Up to 1.5 TB
Aggregate Memory Bandwidth Up to 64 TB/s
Aggregate NVLink Bandwidth 14.4 TB/s
FP4 Tensor Core 144 petaFLOPS2
FP8 Tensor Core 72 petaFLOPS2
INT8 Core 72 petaLOPS22
GPU Memory Up to 192 GB HBM3e per GPU
Decoders/GPU 2x7 NVDEC | 2x7 NVJPEG
Interconnect 5th Generation NVIDIA NVLink: 1.8 TB/s
1Preliminary specifications. May be subject to change.
2With sparsity

Reserve the NVIDIA HGX B200 now

Get ready to build, test, and deploy on The Everywhere Cloud.
Reserve now