Bare Metal GPUs
Get raw, single-tenant access to the latest NVIDIA H100, H200, and AMD MI300X GPUs. Purpose-built for serious AI teams training foundational models, running large-scale inference, and pushing the boundaries of what's possible.
Now Available: Bare Metal AMD Instinct MI300X GPUs
AMD's Instinct MI300X unleashes cutting-edge compute and memory technologies for best-in-class performance for AI model training, inference, and high-performance computing. Built to handle the most data-intensive AI and high-performance workloads, it ensures seamless data processing, reduces bottlenecks, and enables faster results.
Supermaven's AI code completion tool boosts coding speed and efficiency with high-quality suggestions, in-editor AI chat, and a large context window of up to 1 million tokens. Powered by their advanced LLM, Babble, it's 2.5x larger than their previous model.
Bare Metal GPUs
Compare the key specs of Bare Metal GPUs to understand how each can power your AI applications
High-Performance Bare Metal GPUs for AI/ML Workloads
Power your projects with the flexibility of Bare Metal GPUs
Platform Reliability
Support
Pricing
Availability
Bare Metal GPU Resources
What are Bare Metal GPUs?
GPU Memory bandwidth
Monitoring GPU utilization for Deep Learning
Multi-GPU on raw PyTorch with Hugging Face's Accelerate library
How to maximize GPU utilization by finding the right batch size
The Hidden Bottleneck: How GPU Memory Hierarchy Affects Your Computing Experience
FAQ about Bare Metal GPUs
What are Bare Metal GPUs?
Bare Metal GPUs are dedicated, single-tenant servers equipped with eight high-performance GPUs (such as NVIDIA HGX H100, H200, or AMD MI300X) designed to support intensive AI/ML workloads on bare-metal infrastructure.
Which GPU configurations are available for bare-metal servers?
- NVIDIA HGX H100: 8 GPUs, 640 GB GPU RAM
- NVIDIA HGX H200: 8 GPUs, 1,128 GB GPU RAM
- AMD MI300X: 8 GPUs, 1,536 GB GPU RAM
All configurations include dual Intel Xeon Platinum CPUs, 2,048 GiB system RAM, and 61.44 TiB NVMe storage.
How do Bare Metal GPUs compare to GPU Droplets?
GPU Droplets are virtualized and great for flexible, on-demand GPU workloads but share infrastructure. Bare Metal GPUs provide isolated hardware without virtualization—offering consistent, high-performance, and customizable environments for demanding AI training, inference, or HPC tasks.
What types of workloads are best suited for Bare Metal GPUs?
Large-scale model training and distributed deep learning
High-throughput, low-latency inference serving
Custom orchestration setups (e.g., Kubernetes clusters)
High-performance computing and scientific simulation workloads
Applications needing strict data isolation or compliance guarantees
Where are Bare Metal GPUs available globally?
Currently, Bare Metal GPUs are available in New York (USA) and Amsterdam (Netherlands). Additional regions are expected to launch in the future.
What is the provisioning and billing process for Bare Metal GPUs?
Bare Metal GPUs are provisioned via contract with DigitalOcean sales (not on-demand). Billing is handled through Paperspace, requiring a Paperspace account to manage payments.
What networking and connectivity features are included?
One public IP (via NAT) and a private VLAN IP
Public bandwidth up to 40 Gbps (North/South traffic)
Private VPC bandwidth up to 400 Gbps (East/West traffic)
GPU interconnect with IPv6 support and RDMA speeds up to 3.2 Tbps
Can I install custom software and drivers on Bare Metal GPUs?
Yes—you have root-level control. Servers come preconfigured with recommended OS (Ubuntu 22.04 or 24.04), GPU drivers, CUDA, or ROCm—but you can modify or install any software, driver, or toolkit you need.
What security considerations should I be aware of?
Bare Metal GPUs have no default firewall. It's recommended to use UFW or your preferred firewall to secure the public interface. IPMI access is not provided.
Who benefits most from using Bare Metal GPUs?
Advanced developers needing high throughput, full hardware control, and multi-node orchestration
Enterprises and AI-native businesses running consistent, high-utilization AI workloads, requiring performance predictability, compliance, and isolation
Bare Metal GPUs deliver the high-performance, scalable, and secure infrastructure demanded by AI-first, digital-native enterprises.