Vercel AI SDK
TypeScript toolkit for building AI-powered web apps with streaming and agent support
View on GitHubWhat it solves
Building AI-native web interfaces without fighting the browser's streaming model: the SDK handles SSE, token streaming, and component state so you focus on product instead of protocol.
Overview
The Vercel AI SDK is a provider-agnostic TypeScript library for building AI-native web interfaces. Core primitives: `streamText` and `generateObject` for LLM calls; `useChat` and `useCompletion` hooks for React/Next.js; and an `Agent` abstraction added in 2025 for multi-step tool-calling loops. Works with OpenAI, Anthropic, Google, Mistral, and 20+ other providers through a unified interface.
Key facts
- Language
- TypeScript
- License
- Apache 2.0
- Maturity
- Community canon
- Maintainer
- vercel
- Reviewed
- 2026-06-11
Where to start
The AI package README shows the three layers (core, UI hooks, and RSC) and which to reach for depending on whether you're in a server or client context.
packages/ai/README.mdTechnologies
More in this domain
LangGraph
Stateful multi-agent orchestration for LLM applications
The state management problem in multi-agent systems: how you build agents that pass context across turns without losing the execution thread.
AutoGen
Multi-agent conversation framework from Microsoft Research
How to coordinate multiple specialized agents toward a shared goal without writing low-level orchestration logic by hand.
CrewAI
Role-based multi-agent orchestration for collaborative task execution
The boilerplate problem in multi-agent setup: defining roles, delegation rules, and task routing without writing a custom orchestration graph.