BaseAI

dev local-firstdeploy serverlessBaseAIThe first Web AI Framework.The easiest way to build serverless autonomous AI agents with memory. Start building local-first, agentic pipes, tools, and memory. Deploy serverless with one command.Get StartedLearn BaseAI⌘ ~npx baseai@latest initagentic ( pipes | tools | memory ) BASE AI --- DocumentationFind something...Ctrl KLearnStar us on GitHub BaseAI Docs BaseAI is the first web AI framework for building Serverless AI agents with Node.js and TypeScript. OPEN: BaseAI is free and open-source LOCAL: world-class local developer experience CLEAN: unlike existing frameworks, BaseAI has zero bloatware no boilerplate code COMPOSABLE: build composable AI pipes (agents), tools (self-healing), and memory (RAG) SERVERLESS: prod-ready, easily deploy to Serverless AI cloud with npx baseai deploy command BaseAI is the first web AI framework for building Serverless AI agents with Node.js and TypeScript. OPEN: BaseAI is free and open-source LOCAL: world-class local dev experience CLEAN: zero bloatware no boilerplate code COMPOSABLE: build composable AI pipes (agents), tools (self-healing), memory (RAG) SERVERLESS: prod-ready easily deploy to Serverless AI cloud with npx baseai deploy Developers use BaseAI to develop high-quality AI agents with memory (RAG) using TypeScript and then deploy serverless as a highly scalable API using ⌘ Langbase (creators of BaseAI). ⌘ Learn BaseAI(Pipes, Tools, Memory)★ Star BaseAI on GitHub Start by building local AI agents we call Pipes Then create a local semantic memory (RAG) so your AI agents can chat with your data Why we created BaseAI Built for Web Developers90% AI use-cases are on the web — BaseAI is API, TypeScript, and web first Local developer experienceZero-cost local development, version-control, and complete observability logs Deploy to Serverless AI CloudSeriously easy to deploy: One command serverless deployment on ⌘ Langbase Self-Healing Agentic Tool-CallingAuto tool calls and result handling with self-healing agents to 21% extra reduced hallucinations Full-Stack Memory x RAG Vector StoreVector db's aren't enough, full-spectrum parsing, chunking, attributes, retrieval testing, and similarity search Composable AI: Agentic ( Pipes | Tools | Memory )Like React components or Docker containers, AI pipes are composable agents of AI automation workflows ProductsDescriptionAI Pipes(Agents)Pipe is a serverless AI agent. Your custom-built AI agent available locally and online as an API. Local first, highly scalable, dynamic, and inexpensive when deployed. A new LLM computing primitive called Pipe. Pipe is the fastest way to ship your AI features in production. It's like having a composable GPT anywhere.AI Memory(RAG)Memory is a managed search engine available locally and as an API for developers. Our long-term memory solution has the ability to acquire, process, retain, and later retrieve information. It combines vector storage, RAG (Retrieval-Augmented Generation), and internet access to help you build powerful AI features and products.AI Tools(Self-healing)Agentic AI tools that seamlessly work together with AI pipes. Extend the model capabilities of AI Pipes and AI Memory. Connect multiple AI pipes together via tools. Build truly composable AI agents with memory (RAG). GuidesQuickstart Pipe GuideLearn to create a local AI agentic pipe.Read moreQuickstart RAG GuideBuild local RAG using BaseAIRead moreQuickstart Tools GuideLearn to create local tools for LLMs.Read more --- LearnFind something...Ctrl KDocsStar us on GitHubBuild an agentic AI pipe with tools Learn how to build an agentic AI pipe with tools and memory /learnWelcome to the /learn BaseAI course. We use Node.js as an example but there are several other production-ready example in our open-source repo including Next.js, Remix, and more. Check out the examples to see how to use BaseAI in your project. ★ Star BaseAIon GitHubRead the docs instead In these learn guides, you will learn how to locally: Create a summarization agent AI pipe. Run the AI pipe with configuration and meta settings. Create a weather tool that returns the current weather for a given location. Integrate the tool in the agent pipe. Run the AI pipe with the integrated tool. Create a memory and add documents to it. Embed the memory to generate embeddings for the documents. Integrate the memory with the agent pipe. Run the AI pipe with the integrated memory. Initialize npm Create a directory in your local machine and navigate to it. Run the following command in the terminal: mkdir my-ai-project && cd my-ai-project npm init -y npm install dotenv CopyCopied! This command will create a package.json file in your project directory with default values. It will also install dotenv package to read environment variables from .env file. In the next learn guide, we will use BaseAI to create a summarizer agentic AI pipe.