A complete machine for CI teams

Dedicated physical Macs. Nothing shared.

The “Fly” in VMFly is about getting builds off the ground. Every node is a physical Apple Silicon Mac mini—not a slice of a virtual machine.

Each machine is assigned to one customer. Your team controls the Xcode version, runner setup, and job queue, turning the build environment from a shared platform resource into infrastructure you manage yourselves.

Rent a physical Mac Compare 2 plans

From $20.1/day. Check the console for live availability.

Machine lineup2 configurations
Locations4 Asia-Pacific data centers
Availability365 days a year
Clear resource boundaries, not performance hype

Keep build queues within your team

We don't split a host machine across multiple customers. The point of choosing a physical Mac isn't a promise of a fixed compile time—it's knowing exactly which resources and configuration boundaries shape your builds.

Run simulators on physical hardware

Xcode and simulators run in a physical macOS environment, with no virtualization layer in between. Workflows involving the GPU, display output, and simulator coordination don't need to adapt to how a shared VM exposes resources.

That doesn't mean every test will run faster automatically. Test volume, simulator concurrency, project dependencies, and memory usage still need to be validated against your project.

Validate first: archive and simulator test the same commit.

Control your own indexes and caches

Xcode indexing, dependency installs, DerivedData, and archives all read from and write to disk. With a dedicated machine, those jobs don't compete with other customers for the same disk or physical network interface.

Your own runners still share the node's resources. Don't copy concurrency settings from a shared platform; adjust them gradually based on memory pressure and available disk space.

Measure first: builds with cold and warm caches.

Ship without waiting for another customer

On a shared runner, queues can form before a build even starts. A dedicated node removes cross-customer contention from that part of your workflow, so release jobs can enter a queue your team controls.

That doesn't mean the node can never have a queue. When several branches trigger jobs at once, your pipeline rules determine what runs in parallel, which archive goes first, and when UI tests run.

Set first: release priorities and concurrency limits.

From provisioning to the next handover

Every rental comes with a complete Mac mini

A physical machine is more than a label. It determines how a node is registered, who manages its environment, and what happens to its data when the rental ends. Our service follows the same handover process every time.

  1. 01

    Procurement and provisioning

    We use Apple Silicon Mac mini computers and offer two configurations: M4 and M4 Pro. Nodes are deployed in four Asia-Pacific data centers and run year-round, without relying on an office's power or network.

  2. 02

    Payment and registration

    Once payment is confirmed, we assign a physical machine and initialize its system. Each machine is registered to its current customer. Manage the node, billing, and support tickets through the console; check the console for live availability.

  3. 03

    Connect your CI pipeline

    After signing in, open the remote desktop in your browser or connect over SSH and screen sharing. Install or configure your tools, switch Xcode versions, register your CI runner, then run a complete build to verify dependencies, signing, and artifact paths.

  4. 04

    Rental end and data cleanup

    Service stops when the rental ends. We erase the entire disk before assigning the machine to the next customer. Your code, certificates, and build artifacts remain your responsibility; back them up and export them before your rental ends rather than relying on the node's local disk as your only copy.

Four gateways, one consistent handover

Choose a location that fits how your team works

Singapore, Tokyo, Seoul, and Hong Kong cover common working time zones for Asia-Pacific development teams. Start by considering how you'll connect, then look at the network routes your code, dependencies, and artifacts need to take.

SIN

Singapore

A good option for teams working across Southeast Asia. If you regularly check simulators over remote desktop, start here to test the hands-on experience.

Glide M4 and Cruise M4 Pro available.

TYO

Tokyo, Japan

Well suited to development workflows that connect through Japanese networks. Check both source-code downloads and build-artifact uploads when choosing a location.

Glide M4 and Cruise M4 Pro available.

SEL

Seoul, South Korea

For teams connecting from South Korea and nearby networks. Before interactive debugging, check response times and display quality, then choose your desktop resolution and testing approach.

Glide M4 and Cruise M4 Pro available.

HKG

Hong Kong

An option for teams across East Asia. Choose based on your team's network connection, and verify it in practice rather than relying on geographic distance alone.

Glide M4 and Cruise M4 Pro available.

Unattended builds and remote debugging have different needs

Teams that rely on runners for automatic archiving should focus on dependency downloads, repository access, and artifact transfers. Teams that use the desktop for hours of debugging should pay closer attention to responsiveness and visual smoothness. We don't promise millisecond figures without test data; your experience depends on your network connection.

Data centers and connection tips
Start with where the node fits in your workflow

For teams that need control of their build environment

We primarily serve CI leads managing pipelines for development teams of 3–30 people, as well as teams preparing a release and individuals running experiments. The right fit isn't about job titles; it's about whether you need a physical Mac you can configure and keep running.

Everyday iOS / macOS build teams

Move tests for stable branches, archiving, and scheduled builds to runners your team manages. Glide M4 comes with Apple Silicon M4, 16GB of memory, and a 256GB SSD—enough to start with everyday CI checks for a single app.

Migrate one pipeline first and keep your existing build as a comparison. Once you've confirmed that artifacts match and dependencies are reproducible, move other jobs over gradually.

Check: tool versions, caches, and job concurrency.

Teams adding build capacity for release week

Add a node temporarily to separate UI tests, archiving, or branch validation from everyday jobs. Daily rentals are a practical way to cover a defined release cycle without buying a machine that may sit unused afterward.

A new node won't automatically split your existing pipeline. Configure runner labels, credentials, and job routing first so the right tasks actually run on the new machine.

Check: job routing and data export after release.

Unity builds and personal AI experiments

Cruise M4 Pro comes with Apple Silicon M4 Pro, 64GB of memory, and a 2TB SSD. It's suited to evaluating Unity export projects, archiving multiple targets, or experimenting with local models on Apple Silicon.

More memory and storage don't guarantee compatibility with every model or plugin. Check the runtime, model size, and project dependencies, then validate resource needs with a real workload.

Check: compatibility, peak memory, and artifact size.

Be clear about what's included

Clear boundaries for plans, billing, and responsibilities

A Mac cloud server doesn't mean we'll take over every part of your build process. We provide a physical node you can access and use exclusively. Your team remains responsible for project dependencies, runner registration, signing configuration, and data backups.

Two machine configurations

Glide M4
Apple Silicon M4
16GB memory · 256GB SSD
Cruise M4 Pro
Apple Silicon M4 Pro
64GB memory · 2TB SSD

Other configurations are not listed as available machines. If you need more disk space, check the +1TB SSD and +2TB SSD add-ons when ordering. Optional configurations with Thunderbolt 5 daisy-chaining are shown in the console.

USD billing only

Choose from four billing periods: daily, weekly, monthly, or quarterly. We accept USDT-TRC20 and Visa / Mastercard / Amex (via Stripe) only. Available payment gateways are confirmed by the backend.

We assign a physical machine only after payment is confirmed. For questions about billing periods, add-ons, or payment records, check your order and invoice first; a payment action alone doesn't mean the node is ready.

Support by email and console ticket only

Send presales questions, security reports, and business inquiries to support@vmfly.com. If you already have an account, sign in to the console to submit and track a ticket about a node or billing.

When reporting a technical issue, include the node ID, the time and time zone, steps to reproduce, and a redacted log excerpt. Never send passwords, private keys, or full payment card details.

Sign in to submit a ticket

Compare the 2 plans

Choose a dedicated physical Mac for your pipeline based on chip, memory, SSD, and billing period. From $20.1/day, with 4 Asia-Pacific data centers.

Compare rental plans Choose a node