We've just released version 0.4.0 of GraphRAG-SDK, our open-source toolkit for building RAG (Retrieval-Augmented Generation) applications using graph databases.
We built this because we saw many developers struggling to effectively leverage graph structures in their LLM-powered apps.
GraphRAG-SDK aims to simplify the process of creating RAG systems that can handle complex, interconnected data. It provides a set of tools and abstractions that let you focus on your application logic rather than the intricacies of graph operations or LLM interactions.
Some key features in this release:
- Support for multiple LLM providers (OpenAI, Anthropic, Cohere)
- Improved query planning for more efficient graph traversals
- New utility functions for common RAG operations
- Enhanced documentation and examples
We've put a lot of effort into making the SDK easy to get started with. You can install it with a simple pip install graphrag-sdk and be up and running in minutes. Our documentation includes several quickstart guides and sample applications to help you explore the capabilities.
The SDK is completely free and open-source. We're committed to keeping the core functionality free, but we're exploring options for premium features or hosted solutions to sustain development.
We'd love to hear your thoughts on what would be valuable to you.
We're eager to see what the community builds with GraphRAG-SDK.
If you give it a try, we'd really appreciate your feedback - what worked well, what was confusing, what features you'd like to see next. Feel free to open issues on GitHub or join our Discord for discussions.
Check out the project on GitHub: https://github.com/FalkorDB/GraphRAG-SDK
Top comments (0)