Get started with the

world’s largest privately-held cloud

infrastructure company

Powered by AMD Instinct™ GPUs on Vultr Kubernetes Engine

Run OpenFold3 on Vultr Kubernetes Engine

OpenFold3 is an GPU-accelerated protein structure prediction model that predicts the 3D structure of proteins, nucleic acids, and small molecules from sequence data, running on AMD Instinct™ GPUs for high-throughput structural biology predictions.

Deploy AMD AI Workbench

Start with AMD AI Workbench on VKE. It provisions the shared gateway, GPU Operator, and certificate manager that OpenFold3 requires.

Deploy OpenFold3 on AMD AI Workbench

OpenFold3 VKE Application deploys on your existing AMD AI Workbench cluster, with automatic HTTPS and model hosting included.

Deploy an AMD Inference Microservice

Deploy AMD Inference Microservices from the AIM Catalog on your AMD AI Workbench cluster with OpenAI-compatible endpoints.

Deploy a VS Code Workspace

Launch a VS Code workspace on AMD AI Workbench preconfigured with PyTorch, TensorFlow, and ROCm on AMD Instinct™ GPUs.

One-Click VKE Deployment

Deploy directly from the Vultr Marketplace with no additional infrastructure setup.

AMD GPU Acceleration

Runs on AMD Instinct™ GPUs via AMD Inference Microservices (AIM) for high-throughput predictions.

Flexible MSA Alignment Modes

Choose from No MSA, ColabFold server alignment for higher accuracy, or upload a custom .a3m file for full control over alignment.