Route inventory

Server locations:90+ countries, 210+ routes

Routes are listed by region: country, city, route type and streaming support on every row. How IEPL dedicated, relay and direct routes work, and when each one fits, is explained below the tables, with route picks split by everyday browsing, streaming, AI tools, gaming and work. Coverage: 90+ countries / 210+ routes, no device limit, 7-day money-back guarantee.

No logs· 7-day money-back guarantee· No email address required· Alipay / WeChat Pay / USDT
IEPL dedicatedFixed path, prioritised at peak hours RelayOptimised link on the cross-border leg DirectLowest cost, reaches long-tail cities

Coverage overview

Route inventory: 90+ countries in four regional groups

Routes are organised into four groups — Asia Pacific, North America, Europe and the rest of the world — one table each, listing country/region, city, route type and streaming support. The tables show sample cities from common markets; overall coverage is 90+ countries / 210+ routes, and the full list expands by region in the client and on the routes page of your user panel.

Hong Kong Japan Singapore South Korea United States United Kingdom Germany Netherlands France Italy Australia Brazil

The tables deliberately leave out latency, bandwidth and online-user counts: those numbers shift with your local network, carrier peering and time of day, so anything written on a static page stops being true quickly. Route type plus your own use case is a far more reliable way to judge whether a route suits you.

Routes by region

Asia Pacific, North America, Europe and the rest of the world

A single city can have several routes of different types: dedicated routes follow a fixed path, relay routes use an optimised link on the cross-border leg, and direct routes land straight on the overseas node. Pick the city by what you need it for, then pick the route type; the Streaming column shows how well a route carries streaming playback, and because platform licensing changes, actual playback results take precedence.

Asia Pacific

Dedicated · Relay · Direct
Sample routes in Asia Pacific; scroll sideways on narrow screens
Country/RegionCityRoute typeStreaming
Hong KongHong KongIEPL dedicatedYes
JapanTokyoIEPL dedicatedYes
JapanOsakaRelayYes
South KoreaSeoulRelayYes
SingaporeSingaporeIEPL dedicatedYes
TaiwanTaipeiRelayYes
MalaysiaKuala LumpurDirectPartial
ThailandBangkokDirectPartial
AustraliaSydneyRelayYes
IndiaMumbaiDirectPartial

North America

Dedicated · Relay · Direct
Sample routes in North America; scroll sideways on narrow screens
Country/RegionCityRoute typeStreaming
United StatesLos AngelesIEPL dedicatedYes
United StatesSan JoseRelayYes
United StatesSeattleDirectYes
United StatesNew YorkRelayYes
United StatesDallasDirectPartial
CanadaTorontoRelayYes
CanadaVancouverDirectPartial

Europe

Dedicated · Relay · Direct
Sample routes in Europe; scroll sideways on narrow screens
Country/RegionCityRoute typeStreaming
United KingdomLondonRelayYes
GermanyFrankfurtIEPL dedicatedYes
NetherlandsAmsterdamRelayYes
FranceParisRelayYes
ItalyMilanDirectYes
SpainMadridDirectPartial
SwedenStockholmDirectPartial
SwitzerlandZurichRelayPartial

Other regions

Relay · Direct
Sample routes in other regions; scroll sideways on narrow screens
Country/RegionCityRoute typeStreaming
BrazilSão PauloDirectPartial
ArgentinaBuenos AiresDirectPartial
United Arab EmiratesDubaiRelayPartial
TurkeyIstanbulDirectPartial
South AfricaJohannesburgDirectPartial
A city missing from these tables doesn't mean there is no route there. The full list expands by region in the client, and the count shown in the panel is authoritative; overall coverage is 90+ countries / 210+ routes. When a country has several cities, prefer the one closest to the service you're reaching.

Route type

How IEPL dedicated, relay and direct routes differ

The difference between the three route types isn't simply fast versus slow — it's about which path the traffic takes and who carries the cost. Once that clicks, you can stop trying routes one by one.

IEPL dedicated

Fixed path, highest stability

A dedicated route is a point-to-point path at the carrier level: traffic travels the same channel from entry to exit instead of crossing the public internet. The cross-border leg is the most controllable of the three and holds up best at peak hours; the trade-off is expensive capacity and a smaller set of cities.

Suited to long-lived connections that need stability: video calls, remote work, large file sync, and workloads that need a fixed exit. Dedicated capacity exists mainly in major cities and is usually not offered in long-tail ones.

Cost tier: High
Relay

Optimised cross-border leg, a balance of coverage and cost

The entry node first hands traffic to an optimised cross-border link, then hands off to the exit node overseas. The cross-border leg takes the optimised path while the final leg runs over the public internet — a compromise between coverage and cost.

Suited to most everyday use: web browsing, streaming, AI chats. It covers more countries than dedicated routes and costs less, which makes it the default recommendation.

Cost tier: Medium
Direct

Shortest path, dependent on the public internet

The client connects straight to the overseas exit node with no extra hop in between. The path is the shortest, but the cross-border leg relies entirely on the public internet, so peak-hour fluctuations show up directly in your experience.

Suited to light browsing, backup routes and nearby regions. It is the cheapest way to reach long-tail cities and works well as a supplement to dedicated and relay routes.

Cost tier: Low

The three types aren't substitutes but a division of labour: in the same city, dedicated routes keep core services steady, relay routes widen coverage, and direct routes fill in long-tail regions. The client labels the type on every row of the route list; when switching, work down from dedicated → relay → direct — it's the fastest way to narrow down a problem.

Routes and protocols are two separate things: the route decides which path traffic takes, the protocol decides how it is encapsulated. The client ships six protocols — Shadowsocks, VMess, Trojan, VLESS, Hysteria2 and TUIC — and you can switch and combine them independently. Wrong route, change the route; wrong protocol, change the protocol — change one thing at a time so you can tell which layer is at fault.

Choosing a route

Everyday browsing, streaming and AI tools: how to choose

Choosing a route isn't about picking the top row of a table; decide the use case first, then the city, then the route type. The five scenarios below cover most daily use, each with a main approach and a first troubleshooting step when things stall.

Everyday browsing

Nearest relay, stability first

Web pages, search and social feeds are short, frequent requests — insensitive to a single round trip but sensitive to stability. Pick the nearest relay city (Hong Kong, Tokyo, Singapore) and leave the protocol on default. If one site won't load, switch to another route in the same region before changing route type.

Streaming

Pick cities marked Yes for streaming

4K playback depends on sustained bandwidth and jitter, not on a one-second peak. Choose cities marked Yes in the Streaming column, preferring relay or dedicated routes, and move to a dedicated route when peak-hour buffering hits. Catalogues shift with licensing, so the same title may behave differently in different cities — trying another city beats reconnecting over and over.

AI tools

Long connections — skip direct routes

AI chats are long-lived connections: a single session can run for minutes, and a mid-stream drop means starting over. Use common exits such as the United States, Japan or Singapore with a relay or dedicated route, not a direct one. If a reply stalls mid-generation, switching route and refreshing the page is faster than waiting it out.

Gaming

Distance sets the floor, route type sets the ceiling

Games are more sensitive to packet loss and jitter than to bandwidth: physical distance sets the floor, and route type can only pull the ceiling back up. Prefer dedicated routes with exit cities close to the game servers, and pick Hysteria2 or TUIC, which forward UDP more smoothly. Cross-continent matches can't be saved even by a dedicated route — that's distance.

Work

Fixed exit, long connections first

Remote meetings, cloud documents and code repository sync are all long-lived connections that need reconnecting after a drop. Choose a dedicated or stable relay route and stay in the same city as much as possible so each connection doesn't land on a different exit. For team collaboration, write the route into the team's usage notes and skip the repeated troubleshooting.

The order in one line: route type by use case, city by the service you're reaching, then keep two or three regular routes in the client to rotate. Don't keep just one, and don't start from the first row of the list every time.

FAQ

Switching routes, streaming and billing

The route tables answer which routes exist; the questions below answer how to use them, and what to do when something breaks.

Why are there no latency or bandwidth numbers in the tables?

Latency and bandwidth shift with your local network, carrier peering and time of day, so numbers written on a static page stop being true quickly. This page lists only stable attributes: country, city, route type and streaming support. For real-world speed, connect to two or three routes for the same target in the client and trust what you see.

What does Yes in the Streaming column mean?

It means the route's exit region and type are suitable for streaming playback and will usually get you that region's catalogue. Platform licensing changes often, so availability of the same title varies over time — actual playback is what counts; for cities marked Partial, try first and then decide whether to use them long term.

What if the city I need isn't listed?

The tables show sample cities from common markets; overall coverage is 90+ countries / 210+ routes, and the full list expands by region in the client and on the routes page of your user panel. If you can't find a specific city, start with a route in the same country or a neighbouring region — in most cases the difference is small.

A route won't connect — what should I switch?

First another route in the same region, then the route type (dedicated → relay → direct), then the protocol. If all three steps fail, the problem is most likely your local network or the target service; note the time of day, your region and the client version in a ticket and troubleshooting goes much faster.

Do monthly plans and data packs use the same routes?

Yes. Route resources don't depend on the plan tier: the ¥9.9 monthly plan and the ¥658 data pack see the same route list. What differs is the data allowance and how it resets — monthly plans reset each month on the activation date, while data packs last until used up and never expire. If you upgrade mid-term, the price difference is converted into remaining days.

Is peak-hour slowdown the route or my local network?

Congestion on cross-border links mostly hits at peak hours, and local broadband often fluctuates at the same time. The test is simple: at the same time of day, switch your usual service to a dedicated route and try again. If the dedicated route stays steady while the direct route slows noticeably, the bottleneck is the public cross-border leg — pin your usual services to a dedicated or relay route.

Next steps

Once you've picked a route

Signing up needs only a username and password — no email address. The client runs on Windows / macOS / iOS / Android / Linux. Monthly plans start at ¥9.9 with 60GB included, and there's a 7-day money-back guarantee.