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

Dataclysm

Dataclysm is an open-source repository of guides and demos for using embeddings across PubMed, arXiv and English Wikipedia. It includes FAISS-based title-similarity retrieval, RAG summarization with Hermes-2.5-Mistral-7B, reranking examples and a Streamlit interface for ranked article search.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Dataclysm — 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 Dataclysm when you want concrete notebook and app examples for embedding-based article discovery, retrieval-augmented summaries or reranking across PubMed, arXiv and English Wikipedia.

Who it is for

  • Developers exploring semantic article retrieval and RAG workflows.
  • Researchers or technical users who want to explore ranked, title-related results across the covered article datasets.

Documented uses

  • Retrieve articles with titles similar to a query using embeddings and FAISS.
  • Summarize retrieved examples with a local LLM workflow.
  • Explore related article results in a Streamlit search interface.
  • Compare classical score augmentation with experimental LLM-based reranking.
Before you choose

Requirements & limits

  • The demonstrated RAG retrieval is based on title similarity, rather than a documented full-text search workflow.
  • LLM-based reranking is labeled experimental in the repository.
  • The README suggests Python 3.10 and a fresh conda environment for installing dependencies.
  • The repository is licensed under Apache-2.0.

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

Start the Streamlit article-search demo

Launch the documented Spatial Search Engine interface for exploring datasets and ranked, title-related articles.

Before you start

  • A fresh conda environment; the README suggests Python 3.10.
  • The repository and its requirements.txt dependencies.
  • Streamlit available in the environment.

Steps

  1. Install the repository dependencies with %pip install -r requirements.txt.

  2. Change to the streamlit-demo directory.

  3. Run streamlit run app.py.

Expected outcome

The Streamlit Spatial Search Engine app starts for exploring the Dataclysm datasets and ranked article results.

Guide limitations

  • The documented search is based on article titles.
  • The guide documents launching the app, not a verification of successful execution.
Documentation 1 ↗
Evidence and transparency

Sources & review

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