Explore Vultr Products

Scalable cloud virtual machines with shared CPU resources, available in global data center locations. Designed for web applications, development/testing environments, and small to medium workloads.

Purpose-built cloud instances with dedicated vCPUs optimized for specific workload types. Choose from CPU-optimized plans for web serving and compiling, memory-optimized plans for databases and caching, or storage-optimized plans for big data and analytics.

GPU-accelerated computing instances featuring NVIDIA and AMD GPUs for AI, machine learning, and high-performance computing. Supports large language model inference, fine-tuning, image generation, and real-time rendering.

Single-tenant dedicated physical servers offering complete hardware isolation with no virtualization overhead. Ideal for compute-intensive workloads requiring maximum performance, including high-frequency trading, batch processing, and machine learning model training.

High-performance dedicated vCPU instances combining flexible local NVMe storage with optional block storage for maximum throughput. Engineered for storage-intensive and bandwidth-demanding workloads including databases, distributed systems, and content delivery.

Fully managed Kubernetes Engine for orchestrating containerized applications at scale. Handles control plane management, automatic scaling, and networking. Supports multi-node deployments with configurable node pools, high availability options, and seamless integration with Vultr Load Balancers and Vultr Block Storage.

Serverless Inference API endpoints for running pre-deployed AI models without managing infrastructure. Access a curated catalog of large language models, vision models, and embeddings hosted on Vultr. Built-in support for vector storage with RAG, tool calling, and streaming responses.

Flexible NVMe and HDD block storage volumes that attach to any Cloud Compute instance. Resize capacity on demand, boot directly from a volume, and protect your data with on-demand snapshots.

S3-compatible object storage with predictable, tiered pricing. Files are private by default, accessible directly from Cloud Compute instances, with optional Archival Storage for cost-efficient cold data.

Shared, scalable NFS file storage that multiple Cloud Compute instances can mount at once, on Linux or Windows. Resize your volume as storage needs grow, with no need to detach or reformat.

An NFS-compatible gateway that connects Vultr Bare Metal servers to Vultr File System as a mounted share. Create and manage NFS exports, currently free during the API Pilot Program.

Fully managed PostgreSQL databases with automated backups, high availability, and read replicas. Includes PgBouncer connection pooling and support for multiple PostgreSQL versions.

Fully managed MySQL databases with automated backups, read replicas, and easy scaling. Choose from multiple storage engines and manage users and access directly from the Vultr Console.

Fully managed, Redis-compatible Valkey in-memory data store for caching and fast data access. Includes automated backups, read replicas, and granular access control with ACLs.

Fully managed Apache Kafka® clusters for building real-time streaming data pipelines. Includes Kafka Connect and Schema Registry, with clusters available in 33+ global locations.

Automatic, scheduled backups that protect your Cloud Compute instances against data loss. Choose how often backups run, restore them to any instance, or convert a backup into a permanent Snapshot.

Point-in-time snapshots for cloning or restoring Cloud Compute and Cloud GPU instances on demand. Snapshots span all Vultr regions and can also be created from a RAW image URL.

Private networking to securely connect instances within a region or across your account. Add a NAT Gateway for outbound-only internet access, or peer multiple VPC networks together.

Managed load balancing with configurable health checks, multiple algorithms, and free Auto SSL. Deploy a Global Load Balancer across multiple regions for worldwide traffic distribution.

Free domain record management for any domain you own, powered by Vultr's Anycast nameserver network for fast, reliable resolution. Manage A, CNAME, MX, TXT, and other record types, configure DNSSEC and zone settings, and set up reverse DNS (rDNS) for your instances through the Console, CLI, API, or Terraform.

Centralized firewall rules you can apply across multiple instances at once. Create service-based rules and manage up to 50 rules per group from a single, central location.

Static IPv4 or IPv6 addresses you can move between instances for seamless failover. Reserved IPs never expire and can be converted from an existing instance's IP address.

Managed container registry for storing, managing, and distributing OCI-compliant container images. Supports public and private repositories, robot accounts for programmatic access, Docker Hub proxy caching, and image version management.

Upload and boot custom ISO images to install any operating system or specialized workload on your Vultr instances. Supports custom Linux distributions, specialized applications, and testing deployments.

Centrally manage SSH key pairs for secure, passwordless authentication across your Vultr infrastructure. Automatically inject SSH keys during instance provisioning, rotate credentials without downtime, and control access at the account level.

Automate instance initialization and configuration using shell scripts that execute during first boot or on-demand. Streamline deployment of applications, system packages, configuration management, and custom initialization logic.