Report Abuse

Basic Information

RivetKit is an open-source TypeScript library for building long-lived serverless processes called Rivet Actors that provide durable state, realtime events, and scalable compute. It is designed for developers who need stateful compute without timeouts, letting you define actors with state, actions, and broadcastable events. The README includes a quickstart that shows installing the actor package, creating a counter actor with durable state and realtime broadcasts, wiring a server to get or create actors, and calling actions from clients. The project is self-hostable and supports multiple runtimes and drivers so you can run locally or deploy to edge and cloud providers. It also ships examples demonstrating AI agents, chat rooms, multiplayer games, CRDT sync, rate limiting, and multi-tenant patterns to accelerate real projects.

Links

Categorization

App Details

Features
Long-lived, stateful compute that keeps memory between calls and avoids short lambda timeouts. Durable state with automatic persistence and fast local reads and writes. Built-in realtime support including WebSockets and server-sent events for broadcasting actor events. Portable runtimes and drivers for Node.js, Bun, Redis, file system and memory storage with guides for Cloudflare Workers and other hosts. Integrations and clients for React, Next.js, Rust, Hono, Express, and planned or available adapters for tRPC and authentication. Examples and templates for AI agent, chat room, multiplayer game, CRDT sync, rate limiter, per-user DB, and stream processing. Tools for scaling, fault tolerance, and deploy-time flexibility.
Use Cases
RivetKit reduces infrastructure complexity by providing a simple API to define durable actors, actions, and events so teams can build realtime and stateful applications faster. It offers type-safe action calls and an examples library that covers common patterns like collaborative sync, games, AI agents, and multi-tenant SaaS to shorten development time. The Studio provides live state inspection, a REPL, connection monitoring, and hot reload to speed debugging and iteration. Support for multiple storage drivers and hosting targets enables running the same code locally, on private infrastructure, or at the edge for low latency. Built-in scalability and error recovery simplify operating production workloads that require persistent server-side state.

Please fill the required fields*