Vercel AI SDK Documentation
A TypeScript toolkit for building AI features into web applications, with streaming, structured output, tool calling and a provider abstraction that lets you change models without rewriting. The practical default for AI in a JavaScript stack.
Tool
AI & Machine Learning
Vercel
A few hours
Intermediate
Free
What it is
An open source TypeScript library for AI applications, covering the parts that are tedious to write yourself: streaming responses to the browser, parsing structured output against a schema, tool calling loops, and a single interface across model providers.
Where it earns its place
Streaming and structured output. Both are fiddly to implement correctly, both are needed by almost every AI feature, and getting them from a maintained library rather than hand rolling them is a clear win. The provider abstraction also means swapping models is a configuration change rather than a rewrite.
A note on scope
It is a focused library rather than a framework, so it composes with whatever you already use. Read the core generation and streaming sections first; the framework specific helpers make more sense afterwards.
Best for: TypeScript and React developers adding AI features to an existing product.
Ready to start?
Opens on Vercel in a new tab.
Stuck on something specific?
Writing only gets you so far. If you want an answer to your situation rather than the general case, book a session and we will work through it together. Every session is free; a few slots open each week.
Follow along
New writing, resources and project ideas land here first.