Practical overview
When it fits
Use Raptor when building or researching a Python RAG workflow that needs tree-organized retrieval over long documents and you want to work with the underlying library.
Who it is for
- Developers incorporating hierarchical document retrieval into a Python RAG application.
- Researchers comparing tree-structured retrieval with flat document retrieval.
- Teams that want to adapt summarization, question-answering, or embedding models through custom implementations.
Documented uses
- Organize text documents into a recursive retrieval tree.
- Ask questions using indexed document content.
- Save and reload a constructed tree.
- Connect custom summarization, QA, and embedding models.
Requirements & limits
The repository is public and specifies the MIT license. The documented installation requires Python 3.8 or later.
- The method recursively organizes documents into a tree rather than describing only flat document retrieval.
- The README's default setup uses an OpenAI API key.
- Custom summarization, QA, and embedding models can be integrated by extending the provided base classes.
- The README says advanced configuration documentation is still in progress.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Install Raptor and prepare its default configuration
Set up the library and initialize the documented default configuration.
Before you start
- Python 3.8 or later
- An OpenAI API key for the README's default configuration
Steps
Clone the Raptor repository.
Change into the repository directory and install dependencies from requirements.txt.
Set the OPENAI_API_KEY environment variable.
Import RetrievalAugmentation and initialize it with the default configuration.
Expected outcome
A RetrievalAugmentation instance initialized with the documented default configuration.
Guide limitations
- The guide covers setup and initialization, not a verified execution.
- The README says advanced configuration documentation is a work in progress.
Add text, ask a question, and save the retrieval tree
Follow the README's basic-use sequence for indexing text, submitting a question, and saving the tree.
Before you start
- Raptor installed and initialized
- A text file to add
- An OpenAI API key when using the README's default configuration
Steps
Read text from a file and pass it to add_documents.
Pass a question to answer_question and print the returned answer.
Save the constructed tree to a chosen path with save.
Expected outcome
The documented workflow produces a question-answer result from the indexed text and saves the tree to a path.
Guide limitations
- The README gives example code rather than a guarantee of results for a particular document or question.
Compare options for Retrieval-augmented applications
These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.
| Project | Documented purpose | Platform | Access and requirements |
|---|---|---|---|
| Raptor | Organize text documents into a recursive retrieval tree. | Linux, macOS, Windows | The repository is public and specifies the MIT license. The documented installation requires Python 3.8 or later. |
| Langsearch | Crawl website or local data for a knowledge base. | CLI, Self-hosted, Server | |
| Hybrid Search Rag | Combine document retrieval and language-model generation in a configurable workflow. |
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 14 October 2026
- Original website title
- parthsarthi03/raptor · Official homepage title · retained in its source language