Skip to content

Latest commit

 

History

History
12 lines (7 loc) · 914 Bytes

README.md

File metadata and controls

12 lines (7 loc) · 914 Bytes

Decode-ELM

Decoding cited knowledge from scientific preprint servers. This tool will allow you to ask questions and get answers based on scientific papers. It uses OpenAI's GPT models, and you must have your own API key. Each query is about 10k tokens, which costs about only $0.20 on your own API key, charged by OpenAI. Current version searches on different pre-print servers including arXiv, chemRxiv, bioRxiv and medRxiv. Used libraries: PaperQA and langchain.

🚧Under development🚧

This app is hosted on HuggingFace:

https://huggingface.co/spaces/mehradans92/decode-elm

See this tweet for a demo.