#5 · Primary category: Writing & Content

storm

agentic-rag deep-research emnlp2024 knowledge-curation large-language-models naacl nlp report-generation retrieval-augmented-generation

An LLM-powered knowledge curation system that researches a topic and generates a full-length report with citations.

Project last updated:09/30/25

GitHub Stars

31.2K

Forks

2.9K

Contributors

26

License

MIT

Why we included this project

STORM takes a single topic prompt and turns it into a full research report with inline citations, built by asking questions from several angles, pulling source material for each, and only then drafting the article. That makes it a useful base for teams that need grounded long-form writing instead of a chat-style answer. The pipeline is modular, so you can plug in your own language models and retrieval backends, including grounding on documents you provide, which means it works against internal knowledge bases as well as web search. If you are building deep-research or report-generation features, the Python package and the streamlit demo give you a quick way to prototype your own workflow. The Co-STORM variant adds a human-in-the-loop mode, letting someone steer the research conversation as it runs.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category