Knowledge graphs are infinitely better than vector search for building the memory of AI agents. With five lines of code, you can build a knowledge graph with your data. When you see the results, you'll never go back to vector-mediocrity-land. Here is a quick video:

Cognee is open-source and outperforms any basic vector search approach in terms of retrieval relevance. • Easy to use • Reduces hallucinations (by a ton!) • Open-source Here is a link to the repository:

Here is the paper explaining how Cognee works and achieves these results:

I also published the video on my YouTube channel:

Coral AI is the most powerful AI for documents. See the difference yourself:

Parece la gráfica de Obsidian

Amazing

Totalmente contigo—es ver un knowledge graph en acción y ya no quiero saber nada de 'vector-mediocrity-land'. Cambio de vida para la memoria en IA.

Básicamente si tenes un texto complejo tenes q hacer un XML gigante describiendo las relaciones ….. no veo mucha magia ahí sino muchos if en forma de XML, si tenes una base de datos gigantes no terminas mas de hacer eso

graphs just made vectors look basic

Amazing stuff as usual, Santiago Now, one knows what to use next time

I needed this thank you!