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

LLM Notebooks

LLM Notebooks is a public collection of AI example notebooks on GitHub. Its knowledge-graph notebook chunks document text, uses a locally hosted language model to extract concepts and semantic relationships, aggregates repeated concept pairs into weighted edges, and calculates graph properties for visualization. The repository also describes using a concept graph as a retrieval structure for document question-answeri

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

LLM Notebooks — 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 to adapt a Jupyter workflow for building a knowledge graph from document text with a locally hosted model, rather than use a ready-made hosted Q&A product.

Who it is for

  • Developers exploring notebook-based knowledge-graph construction.
  • Readers who want to extract concepts and relationships from a document corpus using a local language model.
  • Practitioners investigating graph-based retrieval structures for document question-answering.

Documented uses

  • Chunk text and extract concepts and semantic relationships with an LLM.
  • Aggregate relationships between recurring concept pairs into weighted graph edges.
  • Calculate node degree and communities to support graph visualization.
  • Explore a knowledge graph as a retrieval structure for document question-answering.
Before you choose

Requirements & limits

The repository is public and specifies an Apache-2.0 license. The documented workflow requires Ollama installed locally; the notebook needs adapting to generate a graph.

  • The worked example processes a PDF, while the README describes the method as applicable to any text corpus.
  • The README names Mistral 7B OpenOrca for extraction but gives ollama run zephyr in its setup instructions.
  • The README describes graph-based retrieval as a possible use of the resulting graph; it does not document a complete question-answering application.
  • The repository uses Jupyter Notebook, Pandas, NetworkX and Pyvis in addition to Ollama and its example model.

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

Set up Ollama for the notebook workflow

Install Ollama locally and pull and start the Zephyr model named in the repository's setup instructions.

Before you start

  • A local machine
  • Ollama installed locally

Steps

  1. Install Ollama from the Ollama website.

  2. Open a terminal and run ollama run zephyr to pull the model and start the Ollama server.

Expected outcome

Ollama pulls Zephyr and starts its server, completing the setup steps described in the README.

Guide limitations

  • This is a setup guide, not a complete notebook execution recipe.
  • The README separately identifies Mistral 7B OpenOrca as its concept-extraction example; the setup command names Zephyr.
  • The notebook must be adapted to generate a graph.
Documentation 1 ↗
Evidence and transparency

Sources & review

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