Paperly is an AI-powered tool that transforms complex research papers into clear, plain-English explanations and runnable code.
- Summary & Explanation: Breaks down papers into key concepts, assumptions, and results.
- Runnable Notebooks: Automatically generates a Python notebook (Colab compatible) to reproduce experiments or explore concepts.
- Multi-Provider AI: Uses a robust chain of AI providers (Groq Llama 3/3.1, Cerebras, SambaNova) for fast, free, and reliable inference.
- Session Persistence: Saves your analysis locally so you can resume later.
- Interactive Dashboard: Track implementation tasks and download code.
- Frontend: Vite + Vanilla JS (no framework overhead)
- Backend: Node.js + Express
- AI Inference: Groq, Cerebras, SambaNova (via OpenAI-compatible APIs)
- Styling: Custom CSS variables for a clean, dark-mode aesthetic
-
Clone the repo
git clone https://github.com/yourusername/paperly.git cd paperly -
Install dependencies
npm install
-
Set up environment variables Create a
.envfile in the root directory:PORT=3001 GROQ_API_KEY=gsk_... # Optional backups: CEREBRAS_API_KEY=csk-... SAMBANOVA_API_KEY=...
-
Run locally
# Terminal 1: Start Backend npm run server # Terminal 2: Start Frontend npm run dev
-
Open browser Go to
http://localhost:5173
The dist folder generated by npm run build can be deployed to any static host.
The server/index.js needs a Node.js environment. You can deploy the entire repo to Render.com as a Web Service:
- Build Command:
npm install && npm run build - Start Command:
npm run server
Copyright (c) 2026. All Rights Reserved.
This project is proprietary software. Unauthorized copying, modification, distribution, or use of this software is strictly prohibited. See the LICENSE file for details.
