OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

Ragu

Ragu is a Python framework that turns text into a knowledge graph and supports question answering over the resulting information using graph, community-summary, vector and mixed retrieval.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Ragu — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Ragu when you are implementing a text-based knowledge assistant and want retrieval options that include both graph structure and chunk vectors, rather than relying only on nearest-neighbor chunk lookup.

Who it is for

  • Developers building question-answering systems over their own text collections.
  • Teams creating knowledge-graph-assisted search or knowledge assistants.
  • Python developers who want to configure extraction, graph construction, storage and retrieval components.

Documented uses

  • Extract entities and relations from text and organize them as a knowledge graph.
  • Answer local questions by retrieving graph neighborhoods and related source chunks.
  • Answer broad corpus-wide questions using community summaries.
  • Use chunk-vector retrieval alone or combine it with graph-based retrieval.
Before you choose

Requirements & limits

The public repository identifies the project as MIT-licensed and documents installation from PyPI or a local source checkout. Python 3.10 or newer is required.

  • Ragu is a framework for developers to integrate and configure, not a ready-made hosted question-answering application.
  • The documented quickstart configures an LLM and an embedder through an OpenAI-compatible client and requires API credentials for that example.
  • Retrieval choices include local graph neighborhoods, global community summaries, chunk-vector search and a mixed engine.
  • The documentation describes entity and relation extraction using the NEREL ontology, with 29 entity and 49 relation types.
  • The README describes async batching for embeddings and vector queries, alongside rate-limited LLM clients.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install Ragu and prepare for a graph build

Install the Python package and identify the configuration needed to build a knowledge graph from text.

Before you start

  • Python 3.10 or newer.
  • Text files to process.
  • An LLM and embedding configuration; the quickstart uses an OpenAI-compatible client configured with OPENAI_BASE_URL and OPENAI_API_KEY.

Steps

  1. Install from PyPI with the command below, or clone the repository and install the local build with the command below.

    Documented command
    pip install graph_ragu
    Documented command
    uv pip install -e .

  2. For local models, install the optional dependencies with the command below.

    Documented command
    pip install graph_ragu[local]

  3. Configure the client, LLM and embedder, then set the working storage folder and language.

  4. Use the documented quickstart components to read files, choose a chunker and artifact extractor, configure the builder, and call build_from_docs.

Expected outcome

A configured Ragu installation and a documented path to building a knowledge graph from a text collection.

Guide limitations

  • The quickstart requires model and embedding configuration; the example uses API credentials.
  • The recipe does not establish that a particular model provider, environment or corpus has been tested.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
RaguTeam/RAGU · Official homepage title · retained in its source language
How we review sourcesSuggest a correction