Incredible Work (No issue, only props)
Not only does this objectively obliterate every other retrieval method, your release timing is PERFECT for the knowledge base I'm building (planned to use llamaindex/graphrag, ~150k documents with some up to 300 pages). I was just about to build it out with Neo4J & Quadrant - I'm now trying this first, no question.
Thank you for this contribution to the ecosystem, incredible work, and keep it up. This is destined to become mainstream - it is only a matter of time.
I'm #2 on your waitlist and thrilled to witness your platform's progression.
Thanks for your kind words @anongecko . Please go ahead with your trials. Kindly let me know if you need any help. You can create a new Issue in github or write down.
You can try the python notebook to test and compare with RAG https://github.com/FastBuilderAI/memory/blob/main/fastmemory_financebench_sota.ipynb