Genkit
Google's open-source AI SDK and toolkit for building full-stack agentic apps and AI features in TypeScript, Go, Python, and Dart.
AI API Tool

PRODUCT OVERVIEW
AI SDK is Vercel's open-source TypeScript toolkit for building AI-powered applications and agents with React, Next.js, Vue, Svelte, Node.js, and other frameworks. It standardizes integration across 100+ model providers, so developers can switch providers with one line of code instead of rewriting for each API. AI SDK Core offers a unified API for text generation, structured objects, tool calls, and agent building, while AI SDK UI provides framework-agnostic hooks for chat and generative interfaces. Streaming, built-in fallbacks, and AI Gateway compatibility support production behavior, and a harness abstraction runs established agent harnesses such as Claude Code, Codex, and Pi. Developers install the ai package, follow getting-started guides, and start from templates for chatbots, RAG knowledge bases, and SQL agents.
Install the ai package with npm and import functions such as generateText in your TypeScript project.
Select from supported providers like Vercel AI Gateway, OpenAI, Anthropic, or Google, and reference the model in your call.
Use AI SDK Core functions to generate text, structured objects, and tool calls with a unified API.
Use AI SDK UI hooks such as useChat to add streaming chat and generative interfaces to your app.
Begin with official templates for chatbots, RAG knowledge bases, multi-modal chat, or natural-language PostgreSQL.
Building AI chatbots with streaming responses
Creating agents with tool calls and multi-step reasoning
Adding generative user interfaces to React and Next.js apps
Building RAG knowledge base applications
Querying PostgreSQL with natural language
Switching between model providers without rewriting code
It is Vercel's open-source TypeScript toolkit for building AI-powered applications and agents with frameworks such as React, Next.js, Vue, Svelte, and Node.js.
Getting-started guides cover Next.js App Router and Pages Router, Svelte, Nuxt, Node.js, Expo, Angular, and TanStack Start, plus backend examples for Express, Hono, Fastify, and Nest.js.
The site states support for 100+ models and lists providers including Vercel AI Gateway, OpenAI, Anthropic, Google, xAI, Azure, Amazon Bedrock, Groq, Mistral, DeepSeek, and others.
Yes. The AI SDK standardizes model integration so you can switch providers with one line of code.
AI SDK Core is a unified API for generating text, structured objects, tool calls, and building agents. AI SDK UI provides framework-agnostic hooks for chat and generative user interfaces.
Yes. Streaming is built in, providing real-time responses without custom parsing, and the docs include guidance on stopping streams and backpressure.
Yes. Official templates include a Next.js chatbot, an internal knowledge base with RAG, multi-modal chat, semantic image search, and natural-language PostgreSQL.
You can ask the community on the Vercel Community AI SDK section, read the documentation, or use the llms.txt file to query an LLM about the docs.
See the official website for current pricing.
COMMUNITY REVIEWS
No reviews yet.