Build AI chatbots in hours not months – ChatRAG

Build AI chatbots in hours not months by ChatRAG provides a Next.js and TypeScript boilerplate for creating chatbot products with retrieval-augmented generation, or RAG. It combines document retrieval, user accounts, customization and payment integrations in a codebase for AI applications.

The package is aimed at developers and entrepreneurs building chatbot SaaS products. The download is 9.9 MB.

What you can build with ChatRAG

• Deploy complete RAG chatbot infrastructure using Next.js and Supabase
• Implement document parsing with LlamaCloud for PDFs, Word, and Excel
• Build vector search using OpenAI embeddings for accurate retrieval
• Create monetization systems with Stripe and Polar payment integration
• Deploy chatbots to WhatsApp without requiring a Business account
• Access 200+ AI models through OpenRouter for flexible chatbot backends

What is ChatRAG?

ChatRAG is a codebase for building AI chatbot applications around RAG. Instead of relying only on a model’s general knowledge, a RAG application retrieves information from documents and uses it as context for responses.

The stack brings together LlamaCloud for parsing documents, OpenAI embeddings for semantic search and a Supabase HNSW vector index for retrieval. These components form the foundation for a chatbot that can work with material supplied to it.

The Next.js boilerplate can be deployed on your own infrastructure and customized for a chatbot product. It also includes payment integrations for developers who want to charge users for the applications they build.

What does ChatRAG include?

The production-ready codebase is written in TypeScript and uses Next.js for building the chatbot application. Its AI stack is pre-configured around LlamaCloud, OpenAI and Supabase, giving developers a starting point for document-based retrieval rather than requiring each component to be assembled from scratch.

The product features extend beyond the retrieval stack. Authentication manages users, and chat history persists across sessions so conversations can be retained. A customization dashboard lets you change prompts and visual elements, including logos and colors, without editing code.

Multi-tenant support provides separate workspaces for individual users or teams. This is relevant when building a SaaS application or creating chatbot solutions for multiple clients whose data needs to remain isolated.

Stripe and Polar are included for subscription monetization. The WhatsApp integration connects a chatbot without requiring a WhatsApp Business account, while OpenRouter provides access to model options including Claude, GPT-4 and Gemini.

• Next.js and TypeScript codebase for AI chatbot applications.
• LlamaCloud, OpenAI and Supabase stack for document parsing and retrieval.
• Stripe and Polar integrations for subscription payments.
• OpenRouter access to more than 200 AI models.
• WhatsApp integration without a Business account requirement.
• Authentication, user management and persistent chat history.
• Dashboard controls for prompts, logos and colors.
• Isolated workspaces for users and teams.

Who is ChatRAG for?

Developers can use the boilerplate as a foundation for AI chatbot products without building every part of the application themselves. SaaS founders can adapt it for customer support or documentation bots, with the RAG system drawing on documents relevant to the application.

Agencies can use the multi-tenant design when building chatbot solutions for multiple clients, while entrepreneurs can use the integrated payment systems when developing subscription-based products.

The product requires developer skills: JavaScript and React knowledge are needed to deploy and customize the code. It is a code boilerplate, not a no-code or drag-and-drop chatbot builder. API services such as OpenAI and LlamaCloud, as well as hosting, can involve ongoing expenses.

Who created ChatRAG?

ChatRAG was built by a team of AI developers focused on making RAG technology accessible to entrepreneurs. The team’s product combines the retrieval architecture with the application and monetization components needed for chatbot projects.

Frequently asked questions

What does ChatRAG include?

ChatRAG includes a Next.js and TypeScript RAG boilerplate with authentication, persistent chat history, a customization dashboard and multi-tenant workspaces. It also brings together document parsing, semantic retrieval, payment and messaging integrations.

Do I need coding experience to use ChatRAG?

Yes. Deploying and customizing the boilerplate requires JavaScript and React knowledge; it is a code-based product rather than a drag-and-drop chatbot builder.

Which AI models can ChatRAG use?

The OpenRouter integration provides access to more than 200 models, including Claude, GPT-4 and Gemini. Developers can use these options for their chatbot backends.

Can ChatRAG support multiple clients or teams?

Yes. Its multi-tenant features provide isolated workspaces for users or teams, making the boilerplate applicable to SaaS products and multi-client agency projects.

Are there costs involved in running a ChatRAG application?

Using services such as OpenAI and LlamaCloud and hosting the application can incur ongoing costs. Those costs depend on usage.