This repository provides a pre-configured Docker environment for building Generative AI applications, integrating Langchain, Neo4j, and Ollama. It's designed for developers and researchers looking for a quick start to explore RAG (Retrieval Augmented Generation) patterns, knowledge graphs, and LLM interactions, offering multiple demo applications for inspiration and immediate use.

How It Works

The stack leverages Docker Compose to orchestrate multiple services: Ollama for local LLM hosting, Neo4j for knowledge graph and vector storage, and Python applications built with Langchain for orchestrating LLM calls, data loading, and RAG pipelines. This approach simplifies setup and dependency management, allowing users to focus on application logic rather than infrastructure.

Quick Start & Requirements

• Install: docker compose up
• Prerequisites: Docker Desktop (ensure version 4.24.x or later is avoided due to known performance issues). Ollama installation is required for macOS and Windows; it runs as a container on Linux. API keys for OpenAI, AWS, or Google are needed for respective LLM/embedding models.
• Configuration: Create a .env file from env.example and configure LLM, database URIs, and API keys.
• Resources: Local LLM hosting via Ollama can be resource-intensive.
• Links: Introduction blog post

Highlighted Details

• Integrates Langchain for LLM orchestration and RAG.
• Supports local LLMs via Ollama and cloud LLMs (OpenAI, Anthropic, Google).
• Utilizes Neo4j for both vector similarity search and knowledge graph context.
• Includes five demo applications: Support Bot, Stack Overflow Loader, PDF Reader, Standalone Bot API, and a Svelte-based UI.

Maintenance & Community

The project is maintained by Docker. Community interaction channels are not explicitly mentioned in the README.

Licensing & Compatibility

The repository appears to be under a permissive license, but specific details are not provided in the README. Compatibility for commercial use would depend on the underlying licenses of Langchain, Neo4j, and Ollama.

Limitations & Caveats

The README warns of performance issues with Docker Desktop versions 4.24.x. The setup requires careful configuration of environment variables, especially API keys for cloud-based models.