BigHugger
GH Repository · parthsarthi03

raptor

The official implementation of RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval

stars
1,759
30-day movement
+10/day
Related entries
60
Connections
1
llmagentsvector-databasepythonframeworkmachine-learningPythonlanguage-modelclusteringretrieval-augmented-generationretrievalrag

This is the official Python implementation of RAPTOR (Recursive Abstractive Processing for Tree-Organized Retrieval), a retrieval-augmented generation method that organizes documents into a recursive tree structure using clustering and summarization. It ships as a Python package under an MIT license.

Use it when you want to implement or experiment with tree-organized retrieval for RAG rather than flat vector search.

Use it to

  • Build tree-structured retrieval over document collections
  • Reproduce RAPTOR research results
  • Integrate hierarchical retrieval into RAG pipelines
  • Experiment with clustering-based document summarization

For Researchers and developers building RAG systems in Python

Role
rag
Language
Python
Licence
MIT
Forks
235
Open issues
38
Last push
2024-09-03
topicsragretrievalclusteringllmvector-databasepython