01 — INFRASTRUCTURE

WebRTC TURN Server Service Provider

Global TURN network with automatic routing to the nearest server—or choose a specific region for data residency compliance.

Get Started Free

No credit card required

Global Regions

31+

Worldwide coverage

Latency

<30ms

From anywhere

Uptime

99.999%

Uptime SLA

Edge PoPs

100+

Points of presence

TRUSTED BY INDUSTRY LEADERS

Unity ESPN Canadian Space Agency InVision

A Canadian company building real-time software since 2013. 25,000+ apps have used Metered TURN.

02 — API MANAGEMENT

Manage TURN Server with REST APIs

Manage your TURN server infrastructure programmatically with our comprehensive REST API

03 — GLOBAL NETWORK

Global TURN Server Cloud Infrastructure

Enterprise-grade infrastructure spanning 6 continents with automatic geo-routing to the nearest server

metered-global-network.map
All Systems Operational
TURN Server Location
Hover for details

31+

Global Regions

6 continents covered

100+

Edge PoPs

Points of presence

<30ms

Global Latency

Typical worldwide

99.999%

Uptime SLA

Live on the status page

IETF STANDARDS COMPLIANT

Full conformance to WebRTC & TURN specifications

RFC 8489 RFC 8656 RFC 8445 RFC 8839 RFC 5769 RFC 5780 RFC 6062 RFC 5768 RFC 6336 RFC 6544 RFC 5928

Supported transports:

UDP TCP TLS DTLS TURNS

Available on ports:

80 443 3478

Bypass restrictive firewalls

04 — PERFORMANCE

High Performance TURN Servers

High-performance, low-latency infrastructure for demanding real-time applications

05 — PRICING

Pricing for TURN Server

Start free, scale as you grow

★ INCLUDED WITH EVERY TURN PLAN

Realtime Messaging is bundled with your TURN server.

Managed signalling for SDP and ICE exchange, with pub/sub channels and presence. Every plan's price includes a connection and message allowance, with no separate subscription; Growth, Business and Enterprise include progressively larger ones.

FREE TRIAL

500 MB

$0/mo

  • 500 MB monthly TURN usage(ingress + egress)
  • Global TURN server access
  • REST API access
  • Realtime Messaging included100 connections · 100k messages/mo
  • Email support
Start Free

No credit card required

GROWTH

150 GB

$99/mo

  • 150 GB monthly TURN usage(ingress + egress)
  • Overage: $0.40/GB
  • 99.95% uptime SLA
  • Realtime Messaging included1,000 connections · 5M messages/mo
  • All Free features
  • Static IP allowlisting
  • Email support
  • Emergency phone support
Get Started
POPULAR

BUSINESS

500 GB

$199/mo

  • 500 GB monthly TURN usage(ingress + egress)
  • Overage: $0.20/GB
  • 99.99% uptime SLA
  • Realtime Messaging included5,000 connections · 25M messages/mo
  • Priority support
  • Static IP allowlisting
  • Emergency phone support
Get Started

ENTERPRISE

2 TB

$499/mo

  • 2 TB monthly usage(ingress + egress)
    Overage: $0.10/GB
  • 99.999% uptime SLA
  • Realtime Messaging included20,000 connections · 100M messages/mo
  • Dedicated support
  • Static IP allowlisting
  • Emergency phone support
Get Started

REALTIME MESSAGING · INCLUDED WITH EVERY TURN PLAN

Realtime Messaging allowance by plan

Concurrent connections and messages per month, included in the plan price.

PLAN CONCURRENT CONNECTIONS MESSAGES / MONTH OVER THE ALLOWANCE
FreeEvery account 100 100k Hard capNo overage charges
GrowthTURN 150 GB · $99/mo 1,000 5M $2.50 / million messages$0.01 / extra peak connection
BusinessTURN 500 GB · $199/mo 5,000 25M $2.50 / million messages$0.01 / extra peak connection
EnterpriseTURN 2 TB and up 20,000 100M $2.50 / million messages$0.01 / extra peak connection

Paid-plan overage is drawn from your account balance and is on by default. Connection overage is charged once per billing period for each connection your peak concurrency reaches above the allowance. The allowance becomes a hard cap if you switch overage off in the dashboard, or if your balance runs out with auto-recharge off. The free tier is hard-capped. Limits and quotas

Need higher limits, or Realtime Messaging without a TURN plan?

CUSTOM PLAN

Need higher volumes?

Tailored solutions for high-volume usage and enterprise needs.

Custom usage quotas Volume discounts Dedicated account manager Priority technical support Custom integrations & API solutions Flexible billing terms

06 — ADVANCED FEATURES

Global STUN TURN Servers

Enterprise-grade tools for multi-tenancy, automation, and white-label deployments

MULTI-TENANCY

Projects: Isolate Customers & Control Quotas

Create isolated environments for different customers, apps, or environments. Perfect for SaaS platforms reselling TURN or enterprises with multiple applications.

  • Per-project bandwidth quotas (e.g., 10GB for Customer A, 50GB for Customer B)
  • Project-scoped API keys — safer than exposing your main secret key
  • Track usage per project with detailed analytics
  • Configurable quota actions: notify only, disable credentials, or none
View Projects API Documentation
project-management.sh
# Create a project with 50GB quota
curl -X POST \
  https://turn.metered.ca/api/v1/project \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "projectName": "customer-acme",
    "quota": 50000000000,
    "quotaExceededAction": "notify"
  }'

# Get project usage
curl https://turn.metered.ca/api/v1/project/usage \
  -H "Authorization: Bearer PROJECT_API_KEY"
webhook-payload.json
// Webhook payload at 80% quota
{
  "event": "project_quota_80_percent",
  "project": "proj_abc123",
  "projectName": "customer-acme",
  "quota": 500000000000,
  "usage": 400000000000
}

// Verify HMAC-SHA256 signature
// Header: X-Metered-Signature-256
const isValid = verifySignature(
  payload,
  signature,
  projectApiKey
);
AUTOMATION

Quota Alert Webhooks with HMAC Security

Receive real-time notifications when projects approach or exceed their quotas. Automate your billing and alerting workflows with secure, verified webhooks.

  • Automatic alerts at 80% and 100% quota thresholds
  • HMAC-SHA256 signature verification for security
  • JSON payload with project ID, name, quota, and current usage
  • Perfect for automated billing and usage monitoring systems
View Webhooks Documentation
WHITE-LABEL

Custom Domain Support

Use your own domain for TURN servers. White-label your infrastructure with a simple CNAME record while leveraging Metered's global network.

  • Use your own subdomain (e.g., turn.yourcompany.com)
  • Global endpoint for automatic geo-routing to nearest server
  • Region-specific endpoints for data residency requirements
  • Supports TURN over TCP and UDP
View Custom Domain Guide
dns-config.txt
# Global routing (auto geo-routing)
CNAME  turn.example.com  →  global.relay.metered.ca

# Region-specific (data residency)
CNAME  turn-eu.example.com  →  eu.relay.metered.ca
CNAME  turn-us.example.com  →  us.relay.metered.ca

PROTOCOL SUPPORT

UDP TCP TLS (use default domain)

PRIVATE BACKBONE ARCHITECTURE

User London TURN London PRIVATE BACKBONE TURN New York User New York Cross-continent traffic routed via dedicated backbone, NOT public internet

10+ Tbps

Network Capacity

<30ms

Global Latency

99.999%

Uptime SLA

INFRASTRUCTURE

Private Global Backbone Network

TURN servers interconnected via a dedicated high-speed private backbone. Cross-continent traffic routes through our network, not the public internet.

  • Consistently lower latency for international calls
  • Stable performance even when public internet is congested
  • High reliability for mission-critical real-time applications
  • 31+ global regions connected via premium backbone
Learn More About Our Network

07 — FEATURES

Enterprise-Grade Capabilities

Port 80 & 443

Runs on standard ports for maximum compatibility

Firewall Bypass

Tested to bypass most firewall rules including DPI

99.999% Uptime

Enterprise grade reliability with proven track record

TURNS + SSL

Secure connections through DPI firewalls

Free STUN

Unlimited STUN usage included free

TCP & UDP

Supports both TCP and UDP protocols

Dynamic Routing

Automatic routing to nearest server

Production Ready

Add credentials and you're ready to go

Cost Effective

Free STUN and pay-as-you-go TURN pricing

Multi-Tenancy

Separate usage by customer or app

Easy Admin

Usage logs, threshold alerts, billing records

Enterprise Scale

Handle any number of concurrent users

08 — USE CASES

Built for Every Use Case

AI Voice & Video

Stream AI-generated audio and video responses with ultra-low latency for real-time AI interactions.

TURN for AI agents

IoT & Real-Time Data

Efficiently transfer data between IoT devices in real time with ultra-low latency.

Real-Time Surveillance

Secure, high-performance real-time surveillance with TURN over TLS (TURNS).

Live Streaming

Stream live content with minimal latency for real-time viewing experiences.

VoIP & Video Calling

High-quality VoIP and video calling even in NAT-restricted environments.

3D Rendering & AR/VR

Ultra-high-speed connections for 3D rendering, AR, and VR experiences.

Telemedicine

Reliable, low-latency streaming for healthcare and telemedicine applications.

Data Channels

Efficient peer-to-peer data sharing using WebRTC DataChannels.

09 — WHY METERED

Built by WebRTC Experts

Trusted infrastructure since 2013

PREMIUM NETWORK

Ultra Low-Latency Global Network

Metered has a global low latency network of TURN Servers across the globe that offer less than 30ms latency from anywhere and under 0.01% packet loss.

  • Automatic Geo-Routing to nearest location
  • 31+ Global locations across 6 continents
  • Ultra-Low Latency Network Routing
Low latency network
Expert support

EXCEPTIONAL SERVICE

Customer Service by Engineers

Our customer service team includes engineers and developers experienced in WebRTC who can handle your technical and non-technical requests quickly.

  • Dedicated account manager on 50 TB and Custom plans
  • Quick response time
  • Phone, Chat and Email Support

POWERFUL APIS

REST APIs and Webhooks

With Metered TURN Server's robust REST APIs, you gain full control of TURN Server management.

  • API to Create/Update/Delete Credentials
  • Create auto-expiring credentials
  • Retrieve detailed usage metrics
View Documentation
API coding
dashboard.metered.ca
Metered TURN Server dashboard showing plan usage, top-up balance and traffic statistics

DASHBOARD & ANALYTICS

Powerful Management Dashboard and Analytics

Use the Metered TURN Server dashboard to manage user credentials, view detailed usage stats and manage billing information.

  • Manage Credentials from the dashboard
  • View Usage and Overages
  • Control costs with usage limits
  • Real-time analytics and usage insights

10 — SIGNALLING

Signalling is included.

TURN Relay carries the media. Signalling establishes the session. One integration covers both halves of the job: getting your peers to find each other, and carrying the media once they do. Signalling is included with every TURN plan.

Every Metered TURN plan includes Realtime Messaging: a managed WebSocket layer that handles SDP and ICE exchange between peers. Session establishment is provisioned alongside the relay rather than procured separately.

  • SDP and ICE exchange over a managed WebSocket connection
  • Publish/subscribe channels with wildcard routing
  • Direct peer-to-peer messaging, without broadcasting to the channel
  • Presence tracking for authoritative connection state

WHAT YOU DON'T HAVE TO BUILD

A separate signalling supplier Included
A WebSocket tier you operate and monitor Managed
Presence and session state you implement Built in
Available on Every TURN plan

Paid TURN plans carry a bundled Realtime Messaging (Signalling) allowance, and overages are available if you go over it. Accounts without a paid TURN plan use the free tier. See the allowance per plan

11 — CREDENTIAL MANAGEMENT

Set credentials to expire automatically — or revoke them anytime via API.

Credentials are created from your backend with an optional expiry and an optional label. Any credential can be disabled, re-enabled, deleted, or deleted in bulk by label.

01 · CREATE

Server-side, with your secret key

Returns a username, a password, and a credential-scoped API key.

02 · LABEL

Optional label

A free-text string on the credential. Filter the credential list by it, update it, or delete by it.

03 · EXPIRE

Optional expiryInSeconds

The credential auto-disables when it elapses. No minimum, no maximum. Omit it and the credential does not expire.

04 · REVOKE

Disable, re-enable, or delete

Delete by label removes every active credential carrying that label and returns the count.

KEY SCOPES

Three keys, three scopes

secretKey

ACCOUNT · SERVER-SIDE ONLY

Creates credentials, lists them, and manages projects and usage. Never exposed in front-end code.

apiKey

CREDENTIAL · FRONT-END SAFE

Issued with each credential. Works with one endpoint only — Get Credential, which returns that credential's ICE servers array.

projectApiKey

PROJECT · SERVER-SIDE

Stands in for the secret key on project endpoints. Can be regenerated, which invalidates the old one immediately.

GETTING IT TO YOUR APP

Three routes, depending on the expiry

Expiring credentials. Your backend creates one, gets back a username, password and a credential-scoped apiKey, and hands that key to the front end. The browser exchanges it for the ICE servers array. When the deadline passes, the credential auto-disables.

Non-expiring credentials. Nothing needs re-creating, so the front end can hold the scoped key and fetch its ICE servers directly. Your secret key stays on the server either way.

From the dashboard. Every credential has a Show ICE Servers Array button that gives you the finished configuration to copy.

Every route returns the same complete array — STUN, plus TURN over UDP and TCP on ports 80 and 443, and TURNS over TLS.

create-credential.sh
curl -X POST \
  "$API/api/v1/turn/credential?secretKey=$SECRET" \
  -H "Content-Type: application/json" \
  -d '{
    "expiryInSeconds": 14400,
    "label": "customer-acme"
  }'

# Both fields are optional.
# Returns: username, password, apiKey

IN THE DASHBOARD

What the credential list shows

  • Username, and password masked behind a reveal control
  • Label, and the date the credential was created
  • Data relayed over the last 30 days, per credential
  • Last used, and current status
  • Search by label, and sort by newest or oldest
  • Copy the ICE servers array, or open the credential's API key

Listing credentials over the API is available on every plan, the free trial included.

12 — DATA RESIDENCY

Use automatic geo-routing, or set the region per credential.

By default, traffic routes to the server nearest each user. Or name a region: media for your German customers remains in Germany, media for your Canadian customers remains in Canada. One account, one integration.

Region control is applied at the credential level rather than the account level. Name a region when you request that credential's ICE servers and it relays through that region. Omit it and traffic routes to the nearest available edge automatically.

A single account can satisfy conflicting residency obligations concurrently — per tenant, without duplicate accounts, parallel environments or a forked integration to maintain.

Full region list and endpoints
get-ice-servers.sh
# Relay this credential from a named region
curl "$API/api/v1/turn/credentials?apiKey=$KEY&region=eu"

# Omit it for the nearest server to the user
curl "$API/api/v1/turn/credentials?apiKey=$KEY"

# Both return the full ICE servers array.

STATIC IP ALLOWLISTING

Tell your network team exactly what to open

We publish the relay IP addresses for every region, as individual addresses and as ranges, so your network team opens exactly what your policy allows — no guesswork, and nothing opened on faith.

How much they open depends on how you route. Automatic geo-routing can place a session in any region, so the allowlist has to cover every region we operate. Pin a credential to a region and the list narrows to that region's addresses alone.

Available on every plan, the free trial included.

13 — THROUGHPUT

No throttling. No per-allocation ceiling.

Metered does not impose a throughput limit on individual TURN allocations. Available bitrate is governed by the client's own connection.

4K video

High-bitrate streams

A single 4K stream through the relay encounters no platform-imposed limit, and neither does a multi-track session carrying several of them concurrently.

Remote rendering

Frame buffers and immersive sessions

Remote-rendered frame buffers and AR/VR workloads are bandwidth-intensive by design. The relay is not the constraint on them.

AI video

Multi-track inference pipelines

Sessions carrying multiple tracks into an inference pipeline relay at whatever rate the client is able to sustain.

WHY THE COMMITMENT HOLDS

Provisioned capacity, not contention

An uncapped allocation is only meaningful if the underlying network can carry it. Metered operates 100+ edge points of presence across 31+ regions.

100+

Edge PoPs

31+

Regions

0

Per-allocation caps

14 — COST CONTROL

Set a spending limit, enable auto-recharge for always-on service.

Fund the account in advance. Relay usage draws down that balance and cannot exceed it, unless you set a threshold and an amount to top it up automatically.

PREPAID BALANCE

Fund the account in advance

  • Fund in advance — no open-ended payment mandate required
  • Relay usage draws down the balance
  • Spend cannot exceed the balance unless auto-recharge is enabled
  • Add funds manually at any time
  • Usage history and payment history in the dashboard
  • Suits agencies, regulated buyers and any team required to cap exposure in advance

USAGE VISIBILITY

Attribute consumption to a tenant

Attributing a given day's consumption to a specific customer is a single API call. Daily relay usage is available per credential for the preceding three months, with your own labels present on every record.

  • Daily usage per credential, three months of history
  • Quota, usage and overage for the current billing cycle
  • Billing API with itemised overages, renewals and recharges
  • Suitable for internal chargeback and cost-recovery reporting

AUTO-RECHARGE: Always funded. Always on.

Set a threshold and a top-up amount, and the account keeps itself funded. Your relay runs continuously while the balance looks after itself.

The dashboard projects how long the current balance lasts at your present rate of consumption, so a top-up is a decision you make early rather than a surprise.

You set the ceiling; auto-recharge respects it. Your overage rate and every top-up are shown alongside the balance, so nothing about the bill is inferred.

15 — SUPPORT

Emergency Phone Support

Every paid plan includes emergency phone support. Dial in with your PIN and you're talking to the engineers who run the network.

Email support is for everything else, and it is available on every plan including the free trial — integration and configuration questions, credential and API behaviour, billing and payment queries, or anything else that needs a considered answer rather than an immediate one.

AVAILABILITY BY PLAN

Growth Email + phone PIN
Business Email + phone PIN
Enterprise Email + phone PIN
Free trial Email

Email support covers every plan and every kind of question. The free trial otherwise includes the paid feature set, global region access included, with emergency phone support the single exception. Both channels are staffed 24/7 by WebRTC engineers.

16 — ANALYTICS AND OBSERVABILITY

TURN server usage analytics by credential, project and region.

Relay traffic is only useful if you can attribute it. Usage breaks down by credential, project, country and transport — over windows from the last five minutes to the last ninety days.

GEOGRAPHY

See where your users are

Relay traffic by country, with client counts and share of total. Each row carries its own trend line over the selected window.

ATTRIBUTION

Know which credential spent it

Credentials ranked by bytes, with share of total and a sparkline. The same breakdown exists per project, so a tenant's consumption is one glance away.

TRANSPORT

Know how they connected

The split across UDP, TCP and TLS — so you can see how much of your traffic is traversing restrictive networks, and plan for it.

PEAK HOUR

Know when your traffic actually arrives

A thirty-day heatmap plots session activity as day of week against hour of day. The busiest cell is named outright — the day, the hour, and the sessions it carried — and hovering any other cell reads back its own count.

The headline strip carries the same question in bytes: the single hour that relayed the most data in the selected window.

BUSIEST CELL, CALLED OUT BY DAY AND HOUR

PEAK

The hour that relayed the most data in the window.

CREDENTIALS IN USE

Distinct credentials that actually relayed, not merely those issued.

PERIOD

Every figure carries the window it was measured over.

LIVE

Watch sessions as they connect

The overview lists sessions from the last five minutes as they arrive, each with its type, the region that served it and how long it has been running. Confirmation that a deploy is relaying arrives in seconds, not at the next daily rollup.

LAST 5 MINUTES

TIME RANGE

Compare against the window before

Last hour, day, week, month or ninety days — each window compared against the preceding one of equal length, so a change in consumption is visible rather than inferred.

1H 24H 7D 30D 90D

17 — INTEGRATION

TURN server API documentation, for developers and AI coding tools.

Metered publishes full API reference documentation for your engineers, and machine-readable reference files so AI coding assistants generate integrations against current API behaviour rather than recalled training data.

Provide our TURN reference file to Claude Code, Cursor, GitHub Copilot, ChatGPT, GPT Codex or Windsurf and the assistant integrates Metered TURN into your application from current documentation.

OPENAPI SPECIFICATION

An OpenAPI specification is published alongside it: import it into Postman, generate a typed client in your language, or incorporate it into contract testing as part of your release process.

WHAT IS PUBLISHED

Machine-readable reference files 10
Prompt templates by use case 17
OpenAPI specification Published
Reference applications Clonable

DEVELOPER RESOURCES

Everything You Need to Get Started

Comprehensive documentation, guides, and API references to integrate TURN servers into your application quickly.

Need help getting started?

Our support team is available to help you integrate TURN servers into your application.

View All Docs

99.999% Uptime Global TURN Servers. Experience Unmatched Performance.

Enhance your applications with our enterprise-grade TURN server infrastructure. Start free today.

Get Started Free