The Stateful Backbone for Web Apps

Real-time messaging, distributed caching, and durable workflows, all in one Elixir-powered engine. Deploy anywhere, scale everywhere.

The Developer
Interface

REALTIME
CP-CORE-REV3

Instant Velocity

Pub/sub on the BEAM: push to subscribers, detached producers, and steady latency as you add rooms and regions.

Realtime examples
Main.ex/SDK
Live
UTF-8TypeScript / NodeJS

Quick Start

# Install CachePuppy
$
npm i -g cachepuppy
# Initialize Project
$
npm i -g cachepuppy init
# Start the Engine
$
npm i -g cachepuppy start

Same install flow on macOS, Linux, and Windows. Walk the commands above, then finish with cachepuppy start to bring the engine online locally, then plug in the SDK for your stack.

What it does

Everything you need to build next-generation distributed systems, baked into a single engine

Self Hosted

Deploy anywhere. Full control over your infrastructure with on-premise or private cloud orchestration.

In-memory DB

Ultra-fast state management that maintains persistence across cold starts and system restarts.

Realtime Sockets

Native pub/sub broadcasting for instant communication across millions of concurrent connections.

Workflows

Durable state machines that manage complex logic, retries, and long-running business processes.

Scales Effortlessly

Distributed sharding and horizontal scaling designed for massive throughput and zero-latency.

SDKs for all languages

Native support for JS, Python, Rust, Go, and more, so integration stays seamless in any stack.

Works with Everything

Native SDKs and cloud-native integrations for every layer of your modern stack

JS logo
JS
Svelte logo
Svelte
Python logo
Python
Rust logo
Rust
iOS logo
iOS
Flutter logo
Flutter
React logo
React
NextJS logo
NextJS
Java Spring logo
Java Spring
Go logo
Go
Android logo
Android
React Native logo
React Native

What People Are Building

View on GitHub
AI & ML

Powerful AI workflows

Build autonomous agents that can extract, summarize, and translate massive datasets. Use durable state machines to handle multi-minute LLM tasks without losing context or timing out.

Performance

Fast, In-memory apps

Create high-performance session stores or real-time leaderboards. CachePuppy's persistence layer ensures your in-memory data survives restarts without the complexity of external DBs.

Real-time

High throughput real-time apps

From live stock tickers to collaborative whiteboards. Handle millions of events per second with native pub/sub broadcasting that bridges the gap between server state and client UI.

Architecture

Topic-based in-memory state

Isolate application state into granular topics. Perfect for building complex SaaS platforms where each tenant requires a dedicated, secure, and isolated memory space.

Waiting for the Architects

We haven't shipped to the world yet. Here's what we're asking our first testers.

This Could Be You

Staff Architects

For the infrastructure leads

“Did the runtime hold up under your real deployment conditions?”

This Could Be You

Lead DevOps

For the scale chasers

“How did CachePuppy handle your highest-traffic moment?”

This Could Be You

Principal Engineers

For the reliability folks

“Which workflows stopped being fragile after you switched?”

This Could Be You

Systems Leads

For the event-driven crowd

“What was the missing link before CachePuppy?”

v2.4.0-stable
GPL v3.0

Open Source
at Heart

CachePuppy is built by a global community of architects. Zero black boxes, full transparency, distributed for everyone.