5 available nodes · Both models available

Place your Cloud Mac closer to your workflow

MiniRent currently offers five nodes: Singapore, Tokyo, Seoul, Hong Kong, and the US West Coast. Every valid order corresponds to one dedicated physical Mac mini, not a virtual machine. When choosing a node, consider your daily connection location, code repository, build artifact destination, and team distribution.

Singapore Tokyo, Japan Seoul, South Korea Hong Kong US West Coast
NODE ROUTING SHEET Device run sheet
  1. SGSingaporeSoutheast Asia gateway
  2. JPTokyo, JapanJapan and Northeast Asia
  3. KRSeoul, South KoreaKorean development teams
  4. HKHong KongHong Kong and nearby regions
  5. US-WUS West CoastNorth American West Coast and trans-Pacific collaboration
The connection map illustrates node distribution only and does not represent network routing. Actual round-trip latency depends on the local carrier, cross-border links, and access time.
05 Available nodesFour in Asia, one on the US West Coast
02 Available modelsM4 16GB and M4 24GB
365 Year-round operationAll nodes run normally year-round
Node directory

Start by narrowing the choice by daily connection location

All five nodes offer MiniRent M4 Core and MiniRent M4 Plus. These combinations are normally available to order and marked “Available” in the directory. Real-time availability and delivery time are returned by the console.

SG
Both models available

Singapore

Ideal for teams whose primary members are in Singapore, Malaysia, Indonesia, Thailand, and nearby Southeast Asian markets. For remote desktop work, code synchronization, and continuous integration management, start by continuously pinging this node from your local network and compare jitter with the Hong Kong node.

Recommended connection region
Southeast Asia
Common workloads
Xcode builds, Fastlane, remote development
Model lineup
M4 / 16GB / 256GB; M4 / 24GB / 512GB
MiniRent M4 CoreAvailable
MiniRent M4 PlusAvailable
Preselect Singapore node
JP
Both models available

Tokyo, Japan

Built for development teams in Japan and Northeast Asia. For workflows that frequently use the macOS graphical interface, verify Xcode build results, or download large dependencies, Tokyo is usually the first node to test for local Japanese connectivity and also works well as a centralized build location for Northeast Asian teams.

Recommended connection region
Japan and Northeast Asia
Common workloads
iOS development, automated testing, build artifact archiving
Model lineup
M4 / 16GB / 256GB; M4 / 24GB / 512GB
MiniRent M4 CoreAvailable
MiniRent M4 PlusAvailable
Preselect Tokyo node
KR
Both models available

Seoul, South Korea

Built for teams whose primary members and office networks are in South Korea. The Seoul node is suitable for moving local build queues to the cloud while keeping interactive connections short. If the repository and artifact destination are elsewhere, also measure the full time required to fetch dependencies and upload artifacts.

Recommended connection region
South Korea and Northeast Asia
Common workloads
CI runners, parallel builds, remote debugging
Model lineup
M4 / 16GB / 256GB; M4 / 24GB / 512GB
MiniRent M4 CoreAvailable
MiniRent M4 PlusAvailable
Preselect Seoul node
HK
Both models available

Hong Kong

Suitable for developers in Hong Kong and nearby regions, and a candidate node for cross-region collaboration with Southeast Asia. When frequently syncing code, certificates, test packages, or model files, do not rely on a single ping. Record the median, jitter, and transfer throughput over time.

Recommended connection region
Hong Kong and nearby regions
Common workloads
Remote development, artifact uploads, MLX experiments
Model lineup
M4 / 16GB / 256GB; M4 / 24GB / 512GB
MiniRent M4 CoreAvailable
MiniRent M4 PlusAvailable
Preselect Hong Kong node
US-W
Both models available

US West Coast

The US West Coast is offered as a single node rather than split across cities. It serves West Coast North American teams and trans-Pacific projects whose repositories, artifact storage, or collaborators are concentrated in North America. Before Asian team members work remotely, compare interactive latency with the share of background build tasks.

Preselect US West Coast node
Recommended connection region
North American West Coast and trans-Pacific teams
Common workloads
Continuous integration, artifact archiving, asynchronous builds
Model lineup
M4 / 16GB / 256GB; M4 / 24GB / 512GB
MiniRent M4 CoreAvailable
MiniRent M4 PlusAvailable

Interactive latency across the Pacific is usually higher than within the region. For unattended builds, testing, and artifact archiving, however, repository and upload-target location may matter more than the operator’s location.

Connection baseline

Median ping comparison across five nodes

Use the table for initial screening, not as a substitute for your own network tests. Remote desktops depend heavily on round-trip latency and jitter; continuous integration should also measure repository fetches, dependency downloads, and build artifact upload speeds.

14:00–16:00Test windowLocal business hours at each connection location
60 PKTSample size60 ICMP packets sent per group
P50MethodMedian round-trip latency after excluding packet loss
WIREDTest networkWired office broadband, not a mobile network
Median ping in milliseconds from each primary connection location to the five Cloud Mac nodes
Connection location Singapore Tokyo, Japan Seoul, South Korea Hong Kong US West Coast
Singapore 24 ms 78 ms 69 ms 39 ms 168 ms
Tokyo, Japan 79 ms 18 ms 36 ms 52 ms 112 ms
Seoul, South Korea 70 ms 35 ms 17 ms 47 ms 128 ms
Hong Kong 39 ms 51 ms 48 ms 16 ms 145 ms
US West Coast 169 ms 113 ms 129 ms 146 ms 24 ms
How to read the table

For extended graphical-interface sessions, prioritize median operator-to-node latency and jitter from continuous tests. For command-line and unattended builds, higher interactive latency may be acceptable, but continue checking throughput between the repository and artifact destination.

Result boundaries

The figures come from baseline tests using a consistent method. Actual results vary with the local carrier, enterprise network policies, cross-region routing, access time, and packet loss. The same connection location can produce different results.

Selection method

Do not choose a node based only on the lowest ping

A node must support the complete workflow, not just one remote login. Use the four steps below to record the real path and make a more reliable choice between candidates.

  1. 01

    Identify daily connection locations

    List where the members who genuinely need to operate the macOS graphical interface or command line are located. Test candidate nodes continuously during working hours and record the median, maximum, jitter, and packet loss. Do not decide from a single minimum reading.

    Record Member location, network type, usual hours, interaction method
  2. 02

    Confirm the repository location

    Actually run clones, submodule fetches, and dependency restoration on each candidate node. For large repositories or dependency-heavy projects, stable throughput to the repository may matter more than saving a few dozen milliseconds between the developer and node.

    Record Full clone time, incremental fetch time, dependency restoration time
  3. 03

    Measure where build artifacts go

    Identify where archives, test reports, symbol files, and model results are ultimately uploaded. Run one upload test using a near-realistic file size and observe sustained speed instead of relying only on small-file benchmarks.

    Record Artifact size, upload destination, average throughput, retry behavior
  4. 04

    Review the distribution of collaborators

    When multiple people share build results, choose the node that minimizes the total path among primary operators, the repository, and the artifact destination. For trans-Pacific teams focused on asynchronous builds, place the machine closer to the repository and archive destination.

    Record Primary operators, reviewers, CI trigger, result recipients
Recommended decision order

Interactive location → Repository path → Artifact destination → Team distribution

If two nodes offer a similar interactive experience, prioritize the one with shorter end-to-end time and fewer failed retries after a real build. For graphics-intensive workflows, give greater weight to interactive latency and jitter.

The node does not change the device attributes

Whether you choose Singapore, Tokyo, Seoul, Hong Kong, or the US West Coast, every valid order corresponds to a dedicated physical Mac mini. Compute resources and local storage are not shared with other tenants; this is not a shared virtual machine.

View data protection details

Run a real task first

After your first connection, run a complete repository pull, dependency restoration, Xcode build, and artifact upload. Validate the node with end-to-end timing rather than relying only on the network latency table.

View the connection and build guide

Confirm the configuration in a run sheet after choosing a node

Choose MiniRent M4 Core or MiniRent M4 Plus, the rental period, node, and storage add-ons. Orders are settled in USD and support only USDT-TRC20 and Visa / Mastercard / Amex (via Stripe); the actual available gateway is determined by the backend response.