Asset 20 8 2
Does AI recommend your business? Run the free check →

Join 15,000 business owners, marketers and entrepreneurs. The Sunday newsletter you'll be annoyed only arrives once a week.

Article

Ably Alternatives (2026): 10 Realtime Messaging Tools

What can replace Ably? Three kinds of tool: managed services that hold the connections for you (PubNub, Azure Web PubSub, Pusher Channels), servers and libraries you run yourself (Centrifugo, Soketi, Mercure, Socket.IO, AnyCable), and products with realtime built in (Liveblocks, Supabase Realtime). Choose the kind your team can support at 2am when connections spike, then trial two on one real channel.

Realtime features look tiny from the outside: a typing indicator, a live score, a cursor sliding across someone else's screen. Underneath, something has to hold thousands of open connections, deliver every message in the right order and cope when a phone drops off a train's wifi and comes back thirty seconds later. Ably's own pitch is exactly that: delivery, ordering and failover. Those three words are your checklist for whatever replaces it.

If what you really want is in-app chat with ready-made UI rather than raw messaging, you're shopping in a different aisle. The list of chat SDK alternatives is a better place to start.

Start with who holds the connections

Every tool on this page answers one question differently: whose servers keep the sockets open?

Managed pub/sub means someone else's servers. You publish from your backend, clients subscribe, and scaling is the provider's problem. Microsoft says it plainly for its own service: you don't have to operate the underlying realtime infrastructure. This is the nearest like-for-like swap for Ably.

Servers and libraries you run yourself means your servers. A Go server can plug into an existing backend through a publish API, a hub can push one-way updates over Server-Sent Events, and a library can sit inside your own app with WebSocket first and long-polling as the fallback. You get control and a lower bill, and you get the pager. If you're coming from the Pusher protocol, one self-hosted server here speaks it, and there's a separate list of alternatives to that protocol's original service.

Realtime that comes with the product is the third route, and it's easy to overlook. If your app is collaborative, a sync platform with presence, comments and multiplayer already built may save months. If your data lives in a Postgres-backed platform, its realtime layer can push database changes out without a separate messaging service at all.

What changed on realtime shortlists in 2026

PartyKit is the big one. Its own homepage says the hosted platform stopped taking deploys on 16 October 2026 and shut down on 23 October 2026, though the code stays open source. If an older comparison or a colleague's notes still list it as a hosted option, cross it off, or treat it as code you'd host yourself.

The other shift is AI. One long-standing messaging platform now calls itself the control plane and networking layer for AI agents. A collaboration platform calls itself the sync engine for the agentic web. Microsoft's service mentions streaming AI-generated tokens to clients. Strip the marketing and there's a real point underneath: if you're adding a chat assistant to your product, streaming its answer word by word is a realtime job, and it belongs on the same infrastructure decision as everything else here.

Read the limitations page before the feature page

Feature pages all look the same. Limitations pages are where tools tell you the truth. One self-hosted server on this list, for example, documents that its connection stats collection doesn't work and that plain and secure WebSocket are its only transports. That's honest, and for plenty of apps neither matters. But if some of your users sit behind corporate proxies that kill WebSockets, or you rely on connection stats for monitoring, those two lines decide it.

When you read any tool's docs, look for five answers. What transports it supports and what happens when WebSocket fails. Whether messages a client missed while offline are replayed on reconnect, and for how long. Whether presence (who's online now) is built in. Whether message order is guaranteed per channel. And what happens to connected clients when a server node restarts. Ably's own pitch covers delivery, ordering and failover; whatever you move to has to answer all five, or your code will have to.

The bill: managed, self-hosted and in between

Managed services all meter usage, but they meter different things. Some count monthly active users, some count messages a day plus concurrent connections, and some cloud services price per unit per day. Free tiers are common and sized for testing, not launch day. The paid plans and business tiers, Ably's included, are on each vendor's own pricing page.

Those numbers don't compare directly, and that's the trap. Before you put two quotes side by side, translate your own traffic into each meter: how many distinct users touch a realtime feature in a month, how many messages a busy day sends, and how many connections are open at your peak hour. A chat-heavy app with few users looks cheap on a per-user meter and expensive on a per-message one; a dashboard watched all day by a big audience is the opposite.

Self-hosting moves the bill from a vendor to your cloud account: you pay for servers and bandwidth, sized on peak connections, and an open-source library costs nothing to license. Several of the self-hostable servers below also sell paid or managed editions with extra features and support. Those prices change, so take them from each vendor's own page on the day you decide.

The hidden cost of self-hosting is people. Someone has to scale it, patch it and answer the alert when connections spike during a launch. If that someone doesn't exist, the managed bill is cheaper than it looks.

An hour on one test channel

Pick one real feature, ideally the busiest one, and rebuild just that channel in each of your top two tools.

In the first twenty minutes, get a bare web page subscribing to a channel and your backend publishing to it with a plain HTTP call or server SDK. Note how many lines it took and how auth works. In the next twenty, open the page in three browser tabs and on a phone. Fire a burst of 100 numbered messages and check every client received all 100, in order. Then turn the phone's wifi off for thirty seconds, publish ten more messages, turn it back on and see whether the missed ten arrive. In the last twenty, check presence: does each client see the others join and leave, and how fast does a closed tab disappear?

That hour tells you more than a week of reading. A tool that loses messages on reconnect in a demo will lose them in production too.

Mistakes that bite after launch

WebSocket doesn't always get through. Some corporate and hotel networks block or drop long-lived sockets, and a tool with no fallback leaves those users staring at a frozen screen. Know your audience before you pick a WebSocket-only option.

Size on connections, not people. One customer with three tabs open and your app on their phone is four connections, and connection limits are what plans and servers run out of first.

Work with me

Want AI doing the heavy lifting in your marketing?

I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.

A single self-hosted node is fine right up to launch day. Find out how the server shares messages across nodes while you still have time to add a second.

Database-backed realtime is brilliant when your data already lives in that database. When it doesn't, it's one more moving part to keep in sync.

Ten tools for pushing live messages to users

PubNub

A long-standing realtime messaging platform for chat, collaboration, IoT control and event-driven apps, now also pitched as networking for AI agents.

Azure Web PubSub

Microsoft's fully managed service for long-lived client connections and message delivery, including streaming AI-generated tokens out to users.

Pusher Channels

The hosted service that gave the Pusher protocol its name: your backend triggers an event on a channel and web or mobile clients receive it straight away.

Centrifugo

A self-hosted Go server that takes client connections over several transports and gives your existing backend a simple publish API.

Soketi

An open-source WebSockets server built on uWebSockets.js that speaks the Pusher protocol, for teams who'd rather host that style of channel themselves.

Mercure

An open protocol and hub for Server-Sent Events, where publishing an update is a single authenticated HTTP POST from your server.

Socket.IO

A library rather than a service, giving two-way event messaging over WebSocket with HTTP long-polling as the fallback.

AnyCable

A Go realtime server that recovers messages missed during dropped connections, with presence and ordering built in, available as open source, managed or on-premise Pro.

Liveblocks

Sync infrastructure for collaborative apps, bringing presence, comments, notifications and multiplayer editing ready to drop in.

Supabase Realtime

The realtime layer inside Supabase, combining Broadcast messages, Presence tracking and Postgres change listening in the same project as your data.

Where to start, by kind of team

Small team, no one keen to run servers, features that look a lot like what you had: start with a managed service. It's the shortest move and the least new risk.

A backend team that already runs its own infrastructure and wants the bill down: a self-hosted server is the strongest fit, as long as you've answered the five questions above about transports, replay, presence, ordering and restarts.

Mostly one-way updates, like dashboards, scores or notifications: Server-Sent Events are simpler than WebSockets and often all you need.

A collaborative product, or data already in Supabase: look at the built-in realtime first. You may not need a separate messaging layer at all.

Do you make a realtime, pub/sub or WebSocket tool? Tool makers can ask to be considered for a featured place on this list.

Want your tool on this list? Featured spots and reviews here are paid partnerships. Here's how to partner with me.

Published and maintained by the Lilach Bullock team, covering marketing, AI and business growth.
Your buyers are asking AI who to use. Does it say you?

See for free whether ChatGPT, Claude, Perplexity, Gemini and Google name you, and get the plan to become the answer.

Check my AI visibility →
Sundays only

Get the Sunday newsletter.

One email a week. AI experiments, marketing tactics, and the workflows Lilach is building right now in her own business.

Subscribe free

Let’s get your marketing running on AI.

Book a free 30-minute call

We figure out what you need, where AI fits in, and what working together would look like.

Book the call →

Or take the 30-second calculator

You’ll see the hours and the money quietly leaking out of your week, and the three workflows worth building first.

Take the calculator →

Or grab the free AI resource library

Prompt packs, templates, checklists, and swipe files. The exact tools I build for paying clients. Yours, free.

Get the library →
Keep reading

More from the blog.