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

Graphrag4Openwebui

Graphrag4Openwebui is a Python API interface that integrates Microsoft Research’s GraphRAG with Open WebUI for retrieval-augmented question answering. It provides local knowledge-base search, global search, Tavily-powered internet search and a combined mode that integrates and ranks results from those methods.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Graphrag4Openwebui — 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 it when you want an Open WebUI setup to query a GraphRAG-backed knowledge base, search broader context, add Tavily web results or combine these retrieval modes.

Who it is for

  • Open WebUI operators adding GraphRAG retrieval to a question-answering setup
  • Developers building search or retrieval-augmented systems around prepared knowledge-base files
  • Teams seeking to use local language or embedding models in the documented configuration

Documented uses

  • Search a local knowledge base using graph-based retrieval.
  • Ask questions that require broader global context.
  • Add internet search results through Tavily.
  • Combine local, global and web retrieval for a broader search response.
  • Connect the resulting API to Open WebUI.
Before you choose

Requirements & limits

The repository is public and specifies an Apache-2.0 license. The documented deployment is self-hosted; setup requires Python 3.8 or higher.

  • Local search expects a prepared input directory with required files such as Parquet files.
  • Tavily web search uses an external Tavily API and its API key.
  • Local LLMs and embeddings are supported through documented configuration variables; the README also shows OpenAI-compatible API settings.
  • The README warns that GraphRAG may need installation from a specific source if the listed dependency installation does not install it.
  • The API exposes chat completions and a model-list endpoint, with Open WebUI configured to use the chat-completions URL.

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 and configure the API

Prepare the Python environment and set the documented configuration before starting the service.

Before you start

  • Python 3.8 or higher
  • The project repository and its requirements file
  • An input directory containing required files, such as Parquet files
  • Model and embedding credentials or local model endpoints as applicable; Tavily API key for Tavily search

Steps

  1. Clone the repository and change into its directory.

  2. Create and activate a Python virtual environment.

  3. Install dependencies with the command below.

    Documented command
    pip install -r requirements.txt

  4. Configure the documented environment variables, including the input directory and the model, embedding, and search settings needed for the chosen setup.

Expected outcome

A configured project environment ready for the documented server-start stage.

Guide limitations

  • The README notes that GraphRAG may need installation from a specific source if the requirements installation fails.
  • This guide covers setup and configuration only; it does not establish that the service was executed or that a particular environment will work.
Documentation 1 ↗
Source-based practical guide

Connect the running API to Open WebUI

Point Open WebUI at the project’s documented chat-completions endpoint.

Before you start

  • A configured GraphRAG4OpenWebUI environment
  • Open WebUI
  • The project API running at the documented local address

Steps

  1. Start the API with the command below.

    Documented command
    python main-en.py

  2. In Open WebUI configuration, set the API endpoint to http://localhost:8012/v1/chat/completions.

  3. Use the documented chat-completions endpoint for search requests; /v1/models is listed for retrieving available models.

Expected outcome

Open WebUI is configured to send search requests to the local GraphRAG4OpenWebUI API.

Guide limitations

  • The README advises that asynchronous operations must be supported by the environment.
  • For large-scale deployment, it recommends considering a production-grade ASGI server.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
win4r/GraphRAG4OpenWebUI · Official homepage title · retained in its source language
How we review sourcesSuggest a correction