Vultr Bare Metal provides dedicated single-tenant infrastructure for high-demand workloads like AI/ML, data analytics, and rendering. Powered by the latest enterprise-grade CPU and GPU hardware, Vultr Bare Metal delivers peak performance and full control. Achieve secure, scalable infrastructure that drives results on your terms.

Download
The flexibility of cloud computing with the raw power of physical servers

Vultr Bare Metal merges the raw power of dedicated servers with the flexibility and scalability of Vultr’s global cloud infrastructure. Offering high-performance single-tenant dedicated servers with AMD and NVIDIA GPUs and top-tier CPUs from AMD and Intel, Bare Metal ensures businesses maintain operational efficiency and technological competitiveness.

No information is required for download

Develop, deploy, and optimize

Cloud-native applications that run every aspect of your business

Instantly discover the applications and services for your business in Vultr Marketplace

Harness a wide variety of plug-and-play SaaS applications to make developing and deploying your cloud applications easier.

Visit Vultr Marketplace

Find the containerized services you need in Vultr Container Registry

Browse the PaaS and SaaS offerings in our marketplace of pre-built Kubernetes containers to accelerate application development, deployment and optimization.

Read the datasheet

Draw on services from our partners to deploy and optimize your cloud applications

Vultr's strategic partnerships with leading IaaS, PaaS, and SaaS providers empower customers to build enterprise-grade cloud solutions without the cost, complexity, or lock-in of hyperscalers.

Learn more about the Vultr Cloud Alliance

Powerful servers for the most demanding workloads

GPU Compute

NVIDIA HGX B200

AI Training and AI Inference
  • 8 x NVIDIA B200 192 GB
  • 2 x 1.92 TB NVMe
  • 8 x 3.84 TB NVMe
  • 2 x AMD EPYC 9575F
  • 128 cores / 256 threads @ 3.3 GHz
  • 3 TB RAM
  • 15 TB Bandwidth
  • 8 x 400 Gbps GPU Network
  • 200 Gbps Network
Starting from
$2.890/GPU/hour

AMD Instinct™ MI325X

AI Training and AI Inference
  • 8 x AMD Instinct™ MI325X 256 GB
  • 2 x 960 GB NVMe
  • 8 x 3.84 TB NVMe
  • 2 x AMD EPYC 9575F
  • 128 cores / 256 threads @ 3.3 GHz
  • 3 TB RAM
  • 15 TB Bandwidth
  • 8 x 400 Gbps GPU Network
  • 100 Gbps Network
Starting from
$2.000/GPU/hour

AMD Instinct™ MI300X

AI Training and AI Inference
  • 8 x AMD Instinct™ MI300X 192 GB
  • 2 x 960 GB NVMe
  • 8 x 3.84 TB NVMe
  • 2 x AMD EPYC 9534
  • 128 cores / 256 threads @ 2.45 GHz
  • 2 TB RAM
  • 15 TB Bandwidth
  • 8 x 400 Gbps GPU Network
  • 100 Gbps Network
Starting from
$1.850/GPU/hour

NVIDIA HGX H100

AI Training and AI Inference
  • 8 x NVIDIA H100 80 GB SXM
  • 2 x 480 GB NVMe
  • 8 x 3.84 TB NVMe
  • 2x Intel Platinum 8480+
  • 112 cores / 224 threads @ 2.0GHz
  • 2 TB RAM
  • 15 TB Bandwidth
  • 8 x 400 Gbps GPU Network
  • 25 Gbps Network
Starting from
$2.300/GPU/hour

NVIDIA GH200

AI Inference
  • 1 x NVIDIA GH200 Superchip 96 GB
  • 1 x 960 GB NVMe
  • 1 x 3.84 TB NVMe
  • NVIDIA Grace CPU
  • 72 Cores
  • 480 GB RAM
  • 15 TB Bandwidth
  • 25 Gbps Network
Starting from
$2.990/GPU/hour

NVIDIA L40S

AI and Visual Computing
  • 8 x NVIDIA L40S 48 GB
  • 2 x 480 GB SSD
  • 4 x 3.84 TB NVMe
  • 2x Intel Gold 6448H
  • 64 cores / 128 threads @ 2.4 GHz
  • 2 TB RAM
  • 15 TB Bandwidth
  • 25 Gbps Network
Starting from
$0.848/GPU/hour

NVIDIA HGX A100

AI Training and AI Inference
  • 8 x NVIDIA A100 80 GB SXM
  • 2 x 960 GB NVMe
  • 8 x 3.84 TB NVMe
  • 2x Intel Platinum 8480+
  • 112 cores / 224 threads @ 2.0 GHz
  • 2 TB RAM
  • 15 TB Bandwidth
  • 25 Gbps Network
Starting from
$1.490/GPU/hour

NVIDIA A100 PCle

AI Training and AI Inference
  • 4 x NVIDIA A100 80 GB PCle
  • 2 x 480 GB SSD
  • 6 x 1.6 TB NVMe
  • 2x Intel Gold 6248R
  • 48 cores / 96 threads @ 2.8 GHz
  • 512 GB RAM
  • 15 TB Bandwidth
  • 25 Gbps Network
Starting from
$1.290/GPU/hour
CPU Compute

Intel E3-1270

$0.179 per hour
  • 2 x 240 GB SSD
  • 4 cores / 8 threads @ 3.8 GHz
  • 32 GB RAM
  • 5 TB Bandwidth
  • 10 Gbps Network
Starting from
$120/month

Intel E-2286G

$0.275 per hour
  • 2 x 960 GB SSD
  • 6 cores / 12 threads @ 4.0 GHz
  • 32 GB RAM
  • 5 TB Bandwidth
  • 10 Gbps Network
Starting from
$185/month

AMD EPYC 4245P

$0.439 per hour
  • 2 x 960 GB NVMe
  • 6 cores / 12 threads @ 3.9 GHz
  • 32 GB RAM
  • 5 TB Bandwidth
  • 25 Gbps Network
Starting from
$295/month

Intel E-2388G

$0.521 per hour
  • 2 x 1.92 TB NVMe
  • 8 cores / 16 threads @ 3.2 GHz
  • 128 GB RAM
  • 10 TB Bandwidth
  • 10 Gbps Network
Starting from
$350/month

Intel E-2288G

$0.521 per hour
  • 2 x 1.92 TB NVMe
  • 8 cores / 16 threads @ 3.7 GHz
  • 128 GB RAM
  • 10 TB Bandwidth
  • 10 Gbps Network
Starting from
$350/month

AMD EPYC 4345P

$0.588 per hour
  • 2 x 1.9 TB NVMe
  • 8 cores / 16 threads @ 3.8 GHz
  • 128 GB RAM
  • 10 TB Bandwidth
  • 25 Gbps Network
Starting from
$395/month

AMD EPYC 7443P

$1.079 per hour
  • 2 x 480 GB SSD
  • 2 x 1.92 TB NVMe
  • 24 cores / 48 threads @ 2.85 GHz
  • 256 GB RAM
  • 10 TB Bandwidth
  • 25 Gbps Network
Starting from
$725/month

AMD EPYC 9254

$1.228 per hour
  • 2 x 480 GB SSD
  • 2 x 1.92 TB NVMe
  • 24 cores / 48 threads @ 2.9 GHz
  • 384 GB RAM
  • 10 TB Bandwidth
  • 25 Gbps Network
Starting from
$825/month

AMD EPYC 9354P

$2.158 per hour
  • 2 x 480 GB SSD
  • 4 x 6.4 TB NVMe
  • 32 cores / 64 threads @ 3.25 GHz
  • 768 GB RAM
  • 10 TB Bandwidth
  • 25 Gbps Network
Starting from
$1,450/month

2 x AMD EPYC 9354

$4.353 per hour
  • 1 x 480 GB NVMe
  • 16 x 6.4 TB NVMe
  • 64 cores / 128 threads @ 3.2 GHz
  • 1536 GB RAM
  • 10 TB Bandwidth
  • 25 Gbps Network
Starting from
$2,925/month

2 x AMD EPYC 7713

$8.185 per hour
  • 2 x 480 GB SSD
  • 10 x 6.4 TB NVMe
  • 128 cores / 256 threads @ 2.0 GHz
  • 2048 GB RAM
  • 25 TB Bandwidth
  • 25 Gbps Network
Starting from
$5,500/month

Develop locally,
deploy globally®

32 cloud data center regions

Amsterdam NL
Test Speed
Atlanta, GA US
Test Speed
Bangalore IN
Test Speed
Chicago, IL US
Test Speed
Dallas, TX US
Test Speed
Delhi NCR IN
Test Speed
Frankfurt DE
Test Speed
Honolulu, HI US
Test Speed
Johannesburg ZA
Test Speed
London GB
Test Speed
Los Angeles, CA US
Test Speed
Madrid ES
Test Speed
Manchester GB
Test Speed
Melbourne AU
Test Speed
Mexico City MX
Test Speed
Miami, FL US
Test Speed
Mumbai IN
Test Speed
New Jersey, NJ US
Test Speed
Osaka JP
Test Speed
Paris FR
Test Speed
Santiago CL
Test Speed
São Paulo BR
Test Speed
Seattle, WA US
Test Speed
Seoul KR
Test Speed
Silicon Valley, CA US
Test Speed
Singapore SG
Test Speed
Stockholm SE
Test Speed
Sydney AU
Test Speed
Tel Aviv IL
Test Speed
Tokyo JP
Test Speed
Toronto CA
Test Speed
Warsaw PL
Test Speed
32 regions
Explore regions

Deploy anything
in seconds
Spin up a new instance with your preferred operating system, Vultr Marketplace application, or a custom ISO in just seconds.

Built for business

Schedule automatic backups, create server snapshots, set up flexible networking, and secure compute instances with on-demand firewall protection.

API and Terraform
Use our API or Terraform provider to quickly create and control your instances.
Global content delivery
Accelerate and secure your content across six continents with Vultr CDN, delivering unmatched speed and accessibility.
DDoS
protection
Our DDoS mitigation system offers protection against layer 3 and layer 4 network attacks.
Bring your own IP space
BGP announcement of your IP space is available in any of Vultr’s worldwide cloud data center locations.

Build a composable cloud with best-in-class infrastructure components and services from Vultr and our partners.

From the infrastructure layer to the platform layer to the application layer, a composable approach gives you ultimate flexibility to assemble the right cloud stack to power your business.

The Strategic Shift to Composable AI-First Cloud Infrastructure

Explore the next frontier of cloud computing: the composable AI stack and its opportunities for organizations.

Read the whitepaper

No information is required for download

Case study
“While big providers offered decent global reach, they didn’t offer comprehensive support, and smaller solutions that offered hands-on service were lacking in locations and global availability.”
Philip Cote
Chief Technology Officer, Edgegap

Read the case study

Bare Metal Simplified™

Powered by fully dedicated, high performance AMD EPYC CPUs, these bare metal servers are ready for big business challenges. Noisy neighbors won't slow you down.

View pricing