Tor relay types describe the different roles that volunteer-run computers play in the Tor anonymity network, and each role handles traffic in a distinct way. Every relay sees only a small part of the path a message takes. Understanding these roles helps readers see why no single operator can easily link a user to a destination.
This article explains the main categories of relays, how a user’s traffic moves through them, and what each relay can and cannot see. It also covers the limits of the design, so readers can judge what Tor protects and what it leaves to other safeguards.
Why Tor Uses Several Kinds of Relays
Tor is a network of volunteer-operated servers called relays. Instead of connecting directly to a website, a user’s software builds a path, called a circuit, through several relays. Each relay knows only the hop immediately before and after it, which limits what any one operator can observe.
Splitting the work into roles is a deliberate design choice. Different relays are suited to different jobs, such as accepting connections from users, passing traffic along, or reaching the public internet. Relays are assigned roles based on their stability, bandwidth and configuration.
The Tor Project publishes a directory of relays and their flags, which the network uses to decide which relay can serve which role. Those flags are the practical basis for the categories described below, and they can change as the network evolves.
The Three Core Roles in a Tor Circuit
A typical Tor circuit has three hops: an entry relay, a middle relay and an exit relay. Traffic is wrapped in layers of encryption, and each relay removes one layer before passing the data on. This layered design is often called onion routing.
The entry relay, sometimes called a guard, is the first hop your computer connects to. It knows your real network address but not where your traffic is going. The middle relay sits between the two and sees neither your address nor the final destination.
The exit relay is the last hop before traffic reaches the open internet. It sees the destination address and any unencrypted content, but it does not know who the original user is. Because of this visibility, exit relays receive particular attention in discussions about risk.
Guard Relays: The First Hop
Guard relays are chosen with extra care because they are the only relays that can see your real network address on a typical circuit. Tor clients keep a small, fairly stable set of guards rather than changing them constantly. The following characteristics are commonly associated with guard relays:
- They are selected from relays that have been stable and reliable for a sustained period.
- They carry enough bandwidth to handle traffic from many users without becoming a bottleneck.
- Clients keep a small set of them over time, which limits exposure to many different entry points.
- They are never used as the exit for the same circuit, keeping the first and last hops separate.
The idea behind guards is to reduce the chance that a user’s traffic enters the network through a malicious or compromised relay.
Because a client keeps the same few guards for a period, an adversary running only a few entry relays has fewer opportunities to observe a given user. Guard selection rules are documented by the Tor Project and are refined over time.
Middle Relays: The Hidden Hop
Middle relays form the core of the Tor network and make up the majority of its capacity. They are not visible to the user’s network address and do not see the final destination. Their job is to forward encrypted traffic from one relay to the next without knowing where it started or where it ends.
Because middle relays hold no special role, they can be run by many volunteers with ordinary connections. This broad base helps keep the network distributed. The more independent middle relays there are, the harder it becomes for a single operator to control a large share of any circuit.
For a reader, it helps to think of middle relays as the anonymous interior of the path. They can decrypt one layer of the message, but that layer only tells them the next relay to forward to, not the origin or the final target.
Exit Relays: The Connection to the Internet
Exit relays are the point where Tor traffic leaves the network and connects to the regular internet. The exit relay opens the connection to the destination website or service on the user’s behalf. Because of this, the exit relay can see the destination and, if the connection is not encrypted with HTTPS, the content as well.
Each exit relay operator publishes an exit policy that lists which ports and destinations the relay will allow. Some exits allow only certain common services, while others allow a broader range. Operators can also document their intent and contact information so that questions about the relay can be answered.
Readers should understand that an exit relay sees traffic as coming from the exit itself, not from the user. Using HTTPS on a website remains important, because it protects the content from the exit relay even though the exit can still see which site was contacted.
Bridge Relays and Pluggable Transports
Bridge relays are relays that are not listed in the public directory, which makes them harder for censors to block. They are used by people in places where access to the public Tor network is restricted. Bridges are a special category because their addresses are distributed selectively rather than published openly.
What Bridge Relays Are For
A bridge is an entry point that is not published in the main Tor directory. Users who cannot reach the public network can obtain a bridge address through a distribution channel and connect to it instead. The bridge then forwards traffic into the normal Tor network.
Because bridges are less visible, they reduce the effect of simply blocking every known relay address. Bridges are commonly used in regions with heavy network filtering, and they serve as a practical workaround rather than a guarantee of access. Their availability depends on the operators who run them and on how well they stay hidden.
How Pluggable Transports Help Bridges Blend In
A pluggable transport is a tool that disguises Tor traffic so it looks like something else to someone watching the connection. Common examples include obfs4, which makes traffic appear random, and WebTunnel, which is designed to resemble ordinary web traffic. These tools are optional and are used together with bridges.
No disguise is perfect, and a determined observer may still identify unusual traffic patterns. Pluggable transports raise the cost of detection rather than removing it. Readers who need this capability should consult the Tor Project’s current documentation for guidance on configuration and expected behavior.
Choosing Whether Bridges Are Right for You
Most people who use Tor do not need a bridge. Bridges are generally relevant when the public network is blocked or when a local network actively interferes with Tor connections. Using a bridge when it is unnecessary adds complexity without a clear benefit.
If you think a bridge may help, check the Tor Project’s official guidance and the guidance of local digital rights organizations in your area. Laws and enforcement vary by country, so it is wise to learn what is permitted where you live before you change how you connect.
How Relay Flags and Directory Authorities Shape Roles
Relays do not assign themselves roles. Directory authorities, a small set of trusted servers operated by independent parties, collect information about relays and publish a consensus document. That document records each relay’s status and flags, which indicate properties such as stability, speed and suitability as a guard or exit.
Clients use the consensus to select relays for each circuit. Because the authorities are run independently, the network does not depend on one party’s judgment. The flags are updated over time as relays come and go, so a relay’s role can change even when its operator does nothing.
For everyday readers, the key point is that role assignment is automated and based on measurable behavior. A new relay usually needs time to prove itself before it receives the more sensitive guard or exit roles, which helps keep the network reliable.
What Each Relay Type Can and Cannot See
Understanding visibility makes the system clearer. An entry relay sees the user’s real address and the fact that Tor is in use, but it cannot read the destination. A middle relay sees only the previous and next hop. An exit relay sees the destination and any unencrypted content, but not the user’s address.
No single relay sees both who is connecting and where they are going. This separation is the core privacy property of the design. It does not make a user invisible, and it does not protect against every kind of observation.
Someone who can watch both ends of a connection may be able to link traffic by timing and volume, a limitation the Tor Project documents openly.
Readers should also remember that Tor protects the network path, not the behavior of the person using it. Logging into an account, sharing identifying details or running outdated software can reveal identity regardless of which relays are used.
Common Misunderstandings About Relay Roles
Several misunderstandings appear often when people first learn about Tor relays. Clearing them up helps set realistic expectations about what the network can do. The following points are frequently raised:
- The exit relay is not the user, and the exit operator cannot directly see who started the connection.
- Using a Tor relay does not make a computer or its owner immune to legal or technical scrutiny.
- Running a relay is not the same as running an exit, and most relays never carry exit traffic.
- Encrypted HTTPS traffic stays protected from the exit relay’s view of the content, though the destination is visible.
Accurate understanding protects both users and volunteers. People who run relays should review the Tor Project’s guidance on operating relays responsibly, including questions about legal exposure and the implications of exit traffic. Those who only use Tor should focus on safe browsing habits that complement the network’s design.
Practical Steps for Learning and Using Tor Safely
If you are new to Tor, start by downloading the official Tor Browser from the Tor Project’s website and keeping it updated. The browser configures the relay path for you, so you do not need to choose relay types manually. Reading the Tor Project’s documentation is the most reliable way to stay current on how roles and defaults change.
Consider a short checklist before relying on Tor for sensitive tasks: confirm you are using the official software, keep the browser updated, use HTTPS sites, avoid logging into personal accounts that identify you, and remember that Tor does not replace good security hygiene. Choosing a strong, unique password and enabling multifactor authentication on important accounts also reduces risk.
Keep your expectations realistic. Tor is a strong tool for reducing network-level tracking, but it works best as one layer among several. Learning how guard, middle and exit relays divide the work is a useful step toward understanding what the network protects and where other safeguards must take over.