Rivet
Rivet is a desktop IDE for creating complex AI agents and prompt chaining, with a TypeScript library (Rivet Core) to run those graphs in your own applications. It supports multiple LLMs and vector databases, and is aimed at developers building AI-powered features.
✨ Key features
- Visual IDE for designing AI agents and prompt chains
- Supports OpenAI GPT-3.5/4, Anthropic Claude, AssemblyAI LeMUR
- Embedding/vector database support for OpenAI and Pinecone
- Speech-to-text integration with AssemblyAI
- Rivet Core TypeScript library for embedding graphs in apps
- Available as prebuilt binaries for Mac, Windows, Linux
🎯 Use cases
- Design and debug complex AI agent workflows visually
- Chain prompts and integrate with external APIs
- Embed AI graphs into your own application using Rivet Core
- Build voice-enabled AI features with AssemblyAI speech-to-text
📦 Installation
🧰 Requirements: Desktop app requires macOS, Windows, or Linux; for Rivet Core, Node.js and npm are needed. API keys for LLM providers (OpenAI, Anthropic, etc.) are required for actual use.
Prebuilt Binaries
Download the latest release for your OS:
For all releases, see the releases page.
Running from Source
See CONTRIBUTING.md for building from source.
🚀 Usage
For the desktop application, download and run the binary. For embedding Rivet Core in your application, refer to the Rivet Integration Getting Started page. The README does not provide a code example.
❓ FAQ
What LLMs does Rivet support?
Rivet supports OpenAI GPT-3.5 and GPT-4, Anthropic Claude Instant, Claude 2, and Claude 3 (Haiku, Sonnet, Opus), and AssemblyAI LeMUR for voice data.
Can I use Rivet graphs in my own application?
Yes, Rivet Core is a TypeScript library available on npm as @ironclad/rivet-core, and @ironclad/rivet-node for Node.js, allowing you to run graphs created in Rivet within your own code.
How do I get help or report issues?
You can join the Rivet Discord server, open an issue on GitHub, or start a discussion for questions and ideas.
Is Rivet free to use?
The README does not mention pricing; it is an open-source project with prebuilt binaries available for download.
📊 Repository
🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.