Create GPU Droplets and choose the image, plan, datacenter, authentication method, and additional options.
How to Create and Configure GPU Droplets
Validated on 26 Sep 2024 • Last edited on 18 Apr 2025
DigitalOcean Droplets are Linux-based virtual machines (VMs) that run on top of virtualized hardware. Each Droplet you create is a new server you can use, either standalone or as part of a larger, cloud-based infrastructure.
How to Use GPU Droplets
In general, you can manage GPU Droplets like non-GPU Droplets, but some features and requirements are specific to GPU Droplets:
Set up automatic mounting for the scratch disk on GPU Droplets.
Install the NVIDIA Data Center GPU Manager (DCGM) and DCGM Exporter to enable health monitoring, diagnostics, and process statistics for NVIDIA GPUs on GPU Droplets.
Enable jumbo frames, which are data packets with larger payloads than standard, on GPU Droplets.
1-Click Models, powered by Hugging Face, let you deploy third-party generative AI models on GPU Droplets with no additional setup and configuration.
This Community tutorial explains how to set up the NVIDIA container toolkit, run Docker for GPU workloads, and install Miniconda to manage Python environments on GPU Droplets.
GPU Droplets vs Bare Metal GPUs
DigitalOcean Bare Metal GPUs and GPU Droplets both provide GPU-based compute resources tailored to AI/ML workloads, but they’re each suited for different use cases. Learn more about the difference between bare metal GPUs and GPU Droplets:
Learn the difference between DigitalOcean Bare Metal GPUs and GPU Droplets to choose the product that suits your use case.