Info Gap

English 中文

An AI daemon for arXiv paper searching. Time and token count is traded for low model requirement, response length and quality.

Getting started

For now, only OpenAI-format API is supported. To get started, first clone the repo:

git clone git@github.com:info-gap/info-gap-server.git

Install these python dependencies:

instructor
arxiv

Create .env and set:

OPENAI_API_URL=<your-api-url>
OPENAI_API_KEY=<your-api-key> # Can be arbitrary if no password

Modify REQUEST in main.py, and run agent:

python main.py