90+ countries / 170+ routes

Global routes and server locations

Routes are grouped into four regions — Asia Pacific, North America, Europe and others. Each route is labeled with one of three line types (IEPL dedicated, relay or direct) plus the streaming services it usually handles. The line type decides how a route performs at peak hours and what it costs, so check the differences before you pick one.

  • IEPL dedicated / relay / direct
  • Unlimited devices online at once
  • 60-day money-back guarantee
Line type legend
IEPL dedicated End-to-end dedicated line with a fixed path
Relay Traffic enters a relay node first, then forwards
Direct The client connects straight to the exit node
90+ countries / 170+ routes · the full list lives in the client
Line basics

How the three line types differ

The difference between IEPL dedicated, relay and direct lines is not the protocol — it is how many hops the data takes from entry to exit, and who maintains those hops. With the same subscription and the same client, switching the line type can change peak-hour performance completely.

1

IEPL dedicated

IEPL is an end-to-end dedicated line: data travels from the entry data center to the exit data center entirely over carrier-leased capacity and never crosses the public internet's international gateways. The path is fixed and jitter is low, which makes it the most stable of the three types — and the most expensive, so it is deployed only in popular exit regions.

Best for
4K streaming, long-lived AI tool sessions, video calls, peak evening hours
Cost
Highest; concentrated in popular exit regions such as Japan, Hong Kong, the United States and Singapore
2

Relay

A relay route first carries traffic to a nearby relay node, which then forwards it to the exit data center. The relay leg usually runs over an optimized path that avoids the most congested public gateways. Pricing sits between dedicated and direct, and this type covers the most countries.

Best for
Everyday browsing, mid-bitrate video, exits in less common regions
Cost
Medium; widest coverage, and the usual way into less common countries
3

Direct

With a direct route the client connects to the exit node with no relay in between. The path is shortest and the cost lowest, but it depends entirely on public international gateways, so jitter at peak hours is more noticeable than with the other two. It suits use cases that are not latency-sensitive, or regions where no dedicated or relay line is available.

Best for
Off-peak browsing and downloads, backup routes in less common regions
Cost
Lowest; suited to light use and backups

It is normal for one region to offer several line types. Popular exit regions (Japan, Hong Kong, the United States, Singapore, Canada) usually carry both IEPL dedicated and relay lines, while less common regions are mostly direct. The client labels the type next to each route name: check the type first, then whether the exit region matches the service you want to reach.

Line type affects the path, not the encryption. All three types use the same encryption and protocol stack, so bank-grade encryption is identical across them. The protocols available in the client (Shadowsocks, VMess, VLESS, Trojan, Hysteria2, TUIC) are independent of line type — changing protocol does not change the route.

To tell which type a route belongs to, read the label next to its name. Routes without a label follow the description inside the client. If a region offers dedicated, relay and direct options, try dedicated first, then relay, then direct — at peak hours that order saves the most time.

Route list

Browse routes by region

The table below groups representative routes into Asia Pacific, North America, Europe and other regions, listing country, city, line type and streaming support. These are sample routes; the full list is in the client.

Asia Pacific
Country / Region City Line type Streaming support
JapanTokyoIEPL dedicatedNetflix · YouTube 4K
JapanOsakaRelayYouTube 4K · Abema
Hong KongHong KongIEPL dedicatedNetflix · Disney+
SingaporeSingaporeIEPL dedicatedNetflix · Disney+
South KoreaSeoulRelayNetflix · Wavve
AustraliaSydneyDirectNetflix · Stan
North America
Country / Region City Line type Streaming support
United StatesLos AngelesIEPL dedicatedNetflix · Hulu · HBO Max
United StatesSan JoseRelayYouTube 4K · Hulu
United StatesNew YorkRelayNetflix · Disney+
United StatesSeattleDirectYouTube 4K
CanadaTorontoRelayNetflix · Crave
CanadaVancouverDirectYouTube 4K
Europe
Country / Region City Line type Streaming support
United KingdomLondonIEPL dedicatedBBC iPlayer · Netflix
GermanyFrankfurtRelayNetflix · Disney+
NetherlandsAmsterdamRelayNetflix
FranceParisDirectNetflix · Canal+
SwitzerlandZurichDirectNetflix
SpainMadridDirectNetflix
Other
Country / Region City Line type Streaming support
United Arab EmiratesDubaiRelayNetflix
TurkeyIstanbulDirectNetflix
IndiaMumbaiRelayNetflix · Prime Video
BrazilSão PauloDirectNetflix
South AfricaJohannesburgDirectNetflix
MexicoMexico CityDirectNetflix

The table shows representative routes per region, not the full set. Routes are added and removed as data centers are maintained and carriers change, so the current list in the client is the final word. After a subscription update, new routes appear in the list automatically — no repurchase and no extra charge.

Picking a route

Choose by use case

A single subscription carries many routes, and picking by use case is faster than testing them one by one. The five scenarios below cover most needs; each one lists the line type and exit region to try first.

1

Everyday browsing and search

Web pages, search and email are not latency-sensitive, but they do need a stable connection. Start with a relay route in a nearby region — Japan or Hong Kong relays in Asia Pacific, for example — and avoid switching frequently once connected. The client's routing rules send local sites direct; only overseas domains go through the route.

First choice: relay · nearby region

2

4K video and streaming

Streaming is sensitive to both bandwidth and exit region. The exit region must match the region your account is registered in, or you will see a region-lock message. Prefer IEPL dedicated lines, then relay lines on optimized paths. If playback buffers, switch to another dedicated line in the same region before considering a different region.

First choice: IEPL dedicated · same region as the streaming service

3

AI tools and long connections

Both the web interface and the API of AI tools are long-lived connections, and streaming output is sensitive to packet loss. Choose a relay or dedicated line with a stable IP and a fixed path, and avoid switching exit IPs repeatedly — signing in to one account from several countries easily triggers risk checks. When calling an API, set the client to global mode or add a separate rule for the API domain.

First choice: relay or dedicated · stable IP

4

Gaming and real-time matches

Games care about round-trip latency and jitter; bandwidth matters less. Pick an IEPL dedicated line in the same region as the game server, with as few hops as possible. Do not pick a route in another region — playing on an Asian server through a US exit stretches latency. Wi-Fi adds jitter of its own, so use a wired connection where you can.

First choice: IEPL dedicated · same region, few hops

5

Remote work and video meetings

Meetings and remote desktops need upload and download at the same time, and a dropped stream hurts more than a slow one. Prefer a dedicated line and add a separate rule for the meeting domains in the client, so meeting traffic does not compete with downloads on the same route. If your company requires a specific exit region, pin the route instead of using auto-select.

First choice: IEPL dedicated · stable upload

Billing and routes

How routes relate to plans

The three monthly subscriptions and the three data packs draw on the same set of routes; there is no plan-specific route pool. What differs is the data allowance and how billing works. Monthly: ¥9.9/month for 60GB, ¥18/month for 250GB, ¥28/month for 500GB, with data resetting every month on the day you started and mid-cycle upgrades prorated over the remaining days. Data packs: ¥158/300GB, ¥358/1000GB, ¥658/3000GB, used until they run out and never expiring.

Every plan allows unlimited devices online at once. One account works on Windows, macOS, iOS, Android and Linux simultaneously, and the route list is identical everywhere. Encryption is bank-grade across the board and does not vary by plan tier; the protocols available in the client are unrelated to the plan as well.

Payment is by Alipay, WeChat Pay or USDT. No email address is needed to sign up — a username and password are enough. The refund promise is 60 days, no questions asked; see the refund policy for the process and payout details.

Note: monthly subscription data resets every month on the day you started, not on the first of the calendar month. Data packs last until used up and never expire. You can hold both at once, and each allowance is counted separately.

Troubleshooting

Can't connect? Check these first

If the routes themselves are fine but the client will not connect, the route is usually not the problem. Work through the five steps below in order — most causes turn up within a few minutes.

  1. Switch line type first. Move from a dedicated line to a relay, or to another route in the same region. If that fixes it, the problem was a single route; if nothing connects, keep going.
  2. Update the subscription. The route list in the client comes from the subscription, and after data center changes an old list may point at entry points that are already offline. Update the subscription manually in the client, then reconnect.
  3. Check the routing rules. In rule mode, traffic to a domain judged local never touches the route at all. Switch to global mode temporarily to verify — if the connection works, the rules are the cause.
  4. Try another protocol. Switching protocol on the same route tells you whether the problem sits in the route or in the protocol layer. The protocols available in the client are Shadowsocks, VMess, VLESS, Trojan, Hysteria2 and TUIC.
  5. Rule out the local network. Turn off other proxy apps and browser extensions, and make sure the system clock is accurate — a large time offset makes the handshake fail. On public Wi-Fi, complete the web sign-in page before connecting the route.

If all five steps fail, note the route name, protocol and error message from the client and send them through a ticket — that speeds up the fix considerably.

Use for Free