This is a Streamlit hosting website, where user can give their own Gemini API key and Qdrant API key, and upload a PDF file and ask any question from that PDF. A interactive user interface is made using streamlit, and the uploaded book content is broken down into multiple chucks and get stored in the Qdrant vector-base. The user questions are processed using gemini-2.0-flash model, where the related content from Qdrant vector-base are retrived and sent along with query to the gemini model.
- Streamlit
- Langchain
- Qdrant
- Google Generative AI
https://chatpdf-genai-z9sddcsazzfxoocx56jtvq.streamlit.app
- Download this repository from github.
git clone https://github.com/Sasikumaran-SA/ChatPDF-GenAI.git
cd ChatPDF-GenAI- Setup the virtual environment. You can setup virtual environment if you want. (Optional)
python -m venv .venv
./.venv/Scripts/activate- Install requirements.
pip install -r requirements.txt- Run the app.
streamlit run app.py- Open the link given above in any browser.
- Give Gemini API key, Qdrant URL, Qdrant API key. (If you don't have one create using the link given below)
- Upload a PDF. (Importantly, the uploading PDF should have text which can be seleted/copied)
- Select Process PDF and wait for some time. This will take time depending upon the size of the pdf and the network speed.
- Now using the dialog box shown below the website, you can ask your questions about the PDF and get the answers from it.