📚 chat-pdf-web

Your AI-powered chatbot for PDFs and web pages – built with Streamlit, LangChain, and Gemini 2.5 Pro.

Live Link 👉 https://chat-with-pdf-or-webpage.onrender.com/

✨ Features

Feature	Description
🗂️ PDF Chat	Drag-and-drop any PDF and start chatting instantly.
🌐 Website Chat	Paste any URL to browse & chat with its content.
🧠 Gemini 2.5 Pro	Powered by Google’s newest reasoning model.
🔍 Vector Store	Fast & persistent embeddings with Chroma.
🖼️ Streamlit UI	Responsive sidebar and chat interface.
🔁 State-ful	Retains chat history and vectorstore across sessions.

🚀 Quick Start

1. Clone

git clone https://github.com/<your-org>/chat-pdf-web.git
cd chat-pdf-web

2. Install

pip install -r requirements.txt
# or
pip install .

Python ≥ 3.10 recommended.

3. Launch

streamlit run main.py

A browser tab will open at http://localhost:8501.

4. Configure

In the sidebar:

Enter your Gemini API key.
Upload a PDF or enter a website URL.
Click Ingest and start chatting!

📖 Usage

Step	UI/CLI
1. Provide key	Sidebar → “Enter your Gemini API key…”
2. Upload PDF	“Upload PDF” file picker.
3. Add website	“Enter website URL” input.
4. Ingest	Click Ingest (downloads, chunk, embed).
5. Chat	Type questions in the chat box.
6. Clear Everything	Click “End Chat” to reset.

🛠 Tech Stack

Layer	Stack
UI	Streamlit, `streamlit-extras`
LLM	Google `gemini-2.5-pro`, `google-generativeai`
Embeddings	`gemini-embedding-001`
Orchestration	LangChain
Vector DB	ChromaDB (persisted)
PDF parsing	PyPDFLoader via `pypdf`
Web scraping	`WebBaseLoader`, BeautifulSoup4
Config	`pyproject.toml` (modern PEP621)

📁 Data Flow

graph LR
A[User inputs PDF/URL] --> B{Ingestion}
B -->|PDF| C[PyPDFLoader]
B -->|URL| D[WebBaseLoader]
C & D --> E[LangChain chunks]
E --> F[Embed with Gemini]
F --> G[Chroma Vector Store]
H[User question] --> I[RAGChatBot.retrieve]
I --> G
G --> J[LLM.answer]
J --> K[Show response in UI]

⚙️ Environment Variables (optional)

Var	Default	Purpose
`GOOGLE_API_KEY`	—	Falls back to sidebar input.
`CHROMA_PERSIST_DIR`	`./chroma_db`	Vectorstore path.
`CHUNK_SIZE`	1000	Text-split parameter.
`CHUNK_OVERLAP`	200	Text-split parameter.

🤝 Contributing

We love community contributions!
Please see CONTRIBUTING.md for guidelines.

TL;DR

Fork.
pre-commit install.
Commit & push on a feature branch.
Create a PR 🚀

🔗 References & Credits

Streamlit – docs.streamlit.io
LangChain – langchain.dev
Google AI – ai.google.dev

MissLostCodes/Chat_with_pdf_or_webpage