Fly.io is a platform that allows developers to deploy applications closer to their users by running them in micro-VMs on a global network of servers. It transforms Docker containers into micro-VMs that can run anywhere in the world, providing better performance through edge computing and simplified deployment processes. The platform specializes in running full-stack applications globally with minimal latency.
Founded
2017
Company Size
51-200 employees
Headquarters
San Francisco, USA
Funding
Series A
REST API for managing Machines, including cloning, scaling, starting, and stopping them programmatically.
Lightweight, secure virtual machines using Firecracker technology that provide strong isolation and fast launch times for applications.
BGP Anycast routes user requests to the nearest data center across 35 regions worldwide for low-latency access.
Transforms Docker containers into micro-VMs deployable anywhere in the world with support for private registries.
Orchestrator for creating, deploying, and scaling apps with framework detection and fly.toml configuration.
Built-in support for globally-distributed Postgres databases with clustering capabilities.
Built-in support for Redis databases integrated with the platform.
Deploy applications to any of 35 global regions with automatic routing to the closest instance.
Attach volumes to Machines for persistent storage beyond ephemeral storage.
Scale Machines horizontally or vertically based on metrics using fly scale command.
WireGuard-based private IPv6 network connecting all VMs across regions seamlessly.
Supports runtimes including Go, JavaScript, Python, Ruby, Elixir, PHP, Rust, .NET, Deno, and Crystal.
Automate deployments directly from version control systems like GitHub.
Tools for managing secrets and environment variables securely via CLI.
Out-of-the-box Prometheus metrics displayed via Grafana dashboard.
Command-line tool for managing apps, Machines, deployments, and configurations.
Common questions about Fly.io features, pricing, and capabilities
Fly.io uses Anycast networking to route user requests to the nearest available data center where your application is running. By deploying your Docker containers across multiple global regions, you ensure that the physical distance between your server and your users is minimized, significantly reducing round-trip times.
Yes, Fly.io provides Fly Volumes, which are local NVMe drives attached to your virtual machines. This allows you to run databases and other stateful applications that require persistent data storage across restarts, unlike traditional ephemeral PaaS environments.
Fly.io offers a simplified way to run Postgres clusters that are integrated into your private network. While it is not a fully managed 'hands-off' service, Fly provides automated tooling for scaling, failover, and backups, giving you the power of a dedicated database with the ease of a PaaS.
Generally, no. If your application is already containerized with a Dockerfile, you can deploy it to Fly.io using the 'fly launch' command. The platform is designed to run standard Docker containers, making it highly compatible with most modern web frameworks and languages.
Fly.io provides a dedicated migration tool and documentation specifically for Heroku users. By using the 'fly launch' command in a Heroku project directory, the CLI can often detect your settings and automatically generate the necessary configuration to move your app to Fly.io.
Absolutely. Fly.io allows you to point your own domains to your application via CNAME or A records. The platform also provides automatic SSL certificate provisioning and renewal through Let's Encrypt, ensuring your global traffic is always encrypted via HTTPS.
Every Fly.io organization gets a private IPv6 WireGuard mesh network. This allows your different applications and databases to communicate securely with each other over an encrypted tunnel without exposing those internal ports to the public internet.
Fly.io charges based on the resources your application consumes, including VM size (CPU and RAM), storage volumes, and outbound data transfer. There are no flat monthly fees for the platform itself; you only pay for the specific compute and bandwidth used by your deployed instances.
Fly.io offers a generous 'Hobby' allowance that covers a small amount of compute, persistent storage, and Anycast IP addresses for free. As long as your usage stays within these specific resource limits, you can run small applications without incurring any monthly costs.
Fly.io isolates applications using Firecracker microVMs, providing strong hardware-level isolation between different customers. Additionally, all internal traffic is routed over an encrypted private network, and persistent volumes can be encrypted at rest to protect sensitive user data.
Fly.io maintains SOC2 Type 2 compliance, which is a standard requirement for many enterprise-level applications. You can request their latest security reports and compliance documentation through their support channels to satisfy your internal auditing requirements.
Fly.io offers extensive documentation, a highly active community forum where engineers frequently participate, and email-based support for customers on paid plans. The community forum is often the fastest way to get technical advice from both Fly staff and other experienced users.
Default usage-based plan for all organizations.
Starting at
$0.00/month
Per dedicated IPv4 address per month
Base price for smallest shared CPU machine (Region: ams)
Provisioned capacity per month
First 10 GBs included
Snapshot storage after 10GB free
Outbound data transfer to public internet
First 10 hostnames included
Single hostname certificates after first 10 free
Standard support package
Starting at
$29.00/month
Monthly support subscription
Premium support package
Starting at
$199.00/month
Monthly support subscription
Enterprise support package
Starting at
$2500.00/month
Starting price for Enterprise support
Managed Kubernetes service
Starting at
$75.00/month
Per cluster per month plus compute/volume costs
User reviews coming soon
We're building our review system to help you make informed decisions.
Performance data coming soon
We're collecting uptime and performance metrics to provide comprehensive insights.