finic-ai/rag-stack

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1586 stars · 138 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

RAGstack is a self-hostable private ChatGPT alternative that deploys open-source LLMs (Llama 2, Falcon, GPT4All) alongside the Qdrant vector database inside your VPC. It includes a server and UI for uploading PDFs and chatting over your organization's knowledge base using retrieval-augmented generation.

Use cases

• deploy a private chatgpt alternative inside my vpc
• chat with my company's internal documents using an llm
• build a corporate knowledge base chatbot with open-source models
• run retrieval augmented generation on-premise for security compliance
• query pdfs and confluence docs with a self-hosted llm
• set up qdrant and llama 2 for enterprise rag

When to choose

• you need data privacy and want LLM inference and embeddings kept inside your own VPC
• you want to avoid OpenAI API costs by using open-source models like Llama 2 or Falcon
• you need a quick, batteries-included RAG stack with vector database, server, and chat UI
• your organization requires on-premise or self-hosted knowledge base search

When to avoid

• you want the highest answer quality from frontier commercial models like GPT-4
• you need a fully managed service without operating Kubernetes/GKE infrastructure
• your project needs active development and frequent updates - the last release was in 2023
• you only need a lightweight embedding or vector search library rather than a full chat application

Facets

application · maturity maintenance

rag llm-inference vector-database chatbot chat-interface self-hosted large-language-models artificial-intelligence self-hosted chatbots self-hosted cloud private-chatgpt vpc-deployment qdrant llama-2 falcon gpt4all knowledge-base pdf-chat corporate-search retrieval-augmented-generation docker web-server typescript

Member repositories

For agents

markdown · JSON · MCP: product_card(name="finic-ai/rag-stack")