From documents to decisions — powered by AI, secured locally.
AskDocs AI is a early stage startup comprising a powerful and private document question-answering app powered by LLaMA 2 and LangChain. Upload your PDF, DOCX, or TXT files, and ask intelligent questions. All data is processed locally to maintain maximum privacy.
- 📄 Supports PDF, DOCX, and TXT files
- 🧠 Local LLaMA 2 model inference (no cloud dependencies)
- 🔍 Intelligent search & similarity-based retrieval
- 📦 Built with Streamlit for an interactive UI
- 🖼️ Custom dark-themed background support
- 🧵 Efficient multi-threaded document parsing
- Python 3.9+
- Download the LLaMA 2 model from HERE!
project_root/
├── gen_ai.py # Streamlit application
├── images/
│ └── freepik__adjust__9850.jpeg
│ └── 767.jpg
├── faiss_index/ # Generated vector store (auto-created)
├── evaluation_logs/ # Generated json file (auto-created)
├── documents/
│ ├── empty.txt
│ ├── sample_doc.docx
│ ├── Advanced_Facts_Octopus.pdf
│ └── sample_txt.txt
└── requirements.txt
# Clone the repo
git clone https://github.com/yourusername/AskDocs_GEN-AI.git
cd AskDocs_GEN-AI
# (Optional) Create a virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install dependencies
pip install -r requirements.txt
# Run
streamlit run app.py
- Upload one or more PDF, DOCX, or TXT documents.
- Click Start The Fun!
- Ask your question in natural language.
- Get answers with cited document sources.
- streamlit
- PyMuPDF
- python-docx
- docx2txt
- requests
- numpy
- langchain
- sentence-transformers
- nltk
- rouge-score
- ctransformers
- faiss-cpu
- huggingface-hub
- You need to download and manually place the LLaMA model.
- Ensure GPU support is configured if needed (set gpu_layers accordingly).
- Image background is customizable — make sure the image path is valid.