This application implements a question-answering chatbot powered by the LLAMA model. Users can upload documents, ask questions about their content, and receive answers. The tool leverages LLAMA embeddings for efficient text processing and offers a user-friendly interface powered by Gradio. The LLM used is "llama-2-7b-chat.Q4_K_M.gguf" found here
- Upload a file containing the text you want to analyze.
- Ask questions about the content of the uploaded file.
- Adjust the chunk size and the k-value according to your preferences.
- Receive answers from the LLM model.
- View the chat history to keep track of previous questions and answers.
Note: Ensure that the document format is supported (PDF, DOCX, TXT).
- Upload a file: Upload the document you want to analyze.
- Ask a question about the content of your file: Enter your question about the uploaded document.
- Chunk size: Set the chunk size for text processing (minimum: 100, maximum: 2048).
- K value: Set the value for k in the retrieval process (minimum: 1, maximum: 20).
- LLM Answer: View the answer generated by the LLM model.
- Chat History: Track previous questions and answers in the chat history.