You handle the code and releases; we deliver and run the physical node. VpsHold provides dedicated cloud Macs, not virtual machines. Start by the day and keep your Mac for as long as your project needs it.
Three configurations, five locations, from $21/day. Connection details are usually ready about four minutes after payment is confirmed. Check the console for live delivery progress and availability.
Already renting? Open a support ticket in the console so the issue stays linked to your order and machine ID.
VpsHold is for individuals and teams who need a macOS development environment. Each service is a real Mac mini, not a slice of a shared host. You can add Mac capacity without buying and managing another machine on-site.
For the duration of your rental, no other customer shares the machine's CPU, memory or storage. Because it's not a virtual machine, another tenant's workload can't compete for resources on the same host.
Dedicated hardware doesn't make every task run at the same speed. Project size, build-cache state, simulator concurrency and network downloads all affect how long your work takes.
Test with a real project before deciding how long to keep it.
The macOS desktop and command line are both available. Use Xcode, simulators, certificates and provisioning profiles, and complete your signing workflow on the physical node.
After delivery, check the Xcode version and project dependencies. If your workflow depends on external devices, network reachability or other hardware, tell us what you need before renting.
A working environment isn't the same as a preconfigured project.
We handle node delivery and hardware faults. You manage the development accounts, code, keys and workloads on the system. For team use, create separate macOS user accounts instead of sharing one set of credentials.
Keep independent backups of important projects. Grant CI keys only the access they need, revoke access when someone leaves the project, and never keep your only copy on a rented node.
A dedicated physical machine is not a backup service.
Choose from Singapore, Tokyo, Seoul, Hong Kong and the US East Coast. These five locations serve development teams across Asia-Pacific and users in eastern North America. For interactive development, consider the operator's network route; for unattended CI, also factor in your code host, dependency downloads and artifact destination.
Well suited to Southeast Asian working hours. A practical option for remote development and testing when the main operator is in the region, and for handovers between teammates in UTC+8.
Choose based on the actual network route from the operator to the node.
For development teams in East Asia. If you need screen sharing, debugging or collaborative testing during Japanese working hours, start by assessing the connection from this location.
Choose based on desktop responsiveness and where the work happens.
For collaboration across Northeast Asia. Sharing a time zone can make scheduling easier for Korean teammates handling day-to-day development or testing, but check the connection in practice.
Choose based on your team's location and working hours.
For teams working across southern China and Southeast Asia. If you use the desktop frequently, test from your own office network rather than assuming proximity guarantees a better connection.
Choose based on reachability from your office network.
For collaboration across eastern North America and Western Europe. When daylight saving time changes, plan cross-time-zone handovers around the current offset rather than assuming a fixed time difference.
Choose based on handover times and where build artifacts need to go.
All three configurations are available in these five locations. The location list doesn't guarantee a fixed network latency or desktop and file-transfer speeds. Check the console for live availability.
All locations operate year-round, 365 days a year. We handle service issues under our monthly uptime commitment. If a service becomes unavailable, we first confirm the scope and duration, then calculate service-time compensation from the recorded outage.
Monthly uptime commitment
99.9%
If uptime falls below 99.9%, we provide service-time compensation based on the duration of the outage.
This is a commitment, not a measurement of uptime on this page. Situations such as actions taken by the user and events beyond our control are covered by the service terms.
This page isn't connected to monitoring records, so we won't label unverified dates as normal. The three consecutive 30-day periods below show the limits of the available data; daily statuses aren't fabricated.
If your connection drops, note when you first noticed it, your machine ID and any error message, then open a ticket in the console. Don't post passwords, private keys or full connection credentials publicly.
Getting a machine online is only the start. We focus on how faults are handled, what happens to data when a node is reclaimed, and how we'll notify you about changes that affect your service. Support is available by email or through tickets in the signed-in console.
Once a physical hardware fault is confirmed, we arrange a replacement based on the circumstances. Before the handover, we verify the order and machine ID; afterwards, we check the connection details and system status.
A replacement doesn't guarantee that projects, keys or local workloads will be restored exactly as they were. Keep independent backups. If you need data moved, specify what to retain in your ticket; never send keys by email.
We don't access files on your machine. After a rental ends, the physical node enters the reclamation process. Its disk is erased before the machine is assigned again, so the next customer won't receive the previous user's environment.
Erasure isn't archiving. Before your rental ends, export your projects, build artifacts and any configuration you need, then revoke repository or CI credentials held on the node. Transfer anything you want to keep before ending the rental.
We'll explain the impact and any action you need to take before changes to delivery, connectivity or the way you use the service take effect. Keep your account email able to receive messages; follow up on related issues through a console ticket.
We handle and communicate unexpected incidents, such as hardware faults, based on the circumstances. We won't present an urgent incident as a routine arrangement or promise an unverified recovery time on a public page.
For pre-sales questions, security reports or compliance documents, email us at support@vpshold.com. For an active instance, please use a console ticket so your order, node and support history stay together.
These first-person scenarios represent three common types of users and can help you decide whether the service fits your needs. They are illustrative examples, not real customer reviews or promises about performance or response times.
“I want my model and dependencies in the same environment. I don't want to rebuild everything after each experiment. First, I'll check that the memory and storage can handle my workload, then decide how long to keep the machine.”
AI researcher · Example scenario
“I need the team to share a consistent project environment, not a shared password. Everyone gets their own system account, builds and handovers follow our schedule, and one person remains responsible for the order.”
Outsourced engineering lead · Example scenario
“I just need an extra Mac before a release. I'll pull the project and run a regression test first, check the toolchain and signing setup, then decide whether to rent by the day or for longer.”
Independent developer · Example scenario
The VpsHold founding team's service principles
When resource ownership, connection details and end-of-rental procedures are clear, you can focus on your project. Choose the right configuration and location. If you have toolchain, migration or team-workflow requirements, tell us before you order.