A collection of hands-on Generative AI projects exploring different LLMs, frameworks, and tools.
This repository demonstrates practical applications of OpenAI, Ollama, LangChain, Streamlit, Groq , huggingface and more.
A conversational chatbot built with Streamlit and LangChain, supporting:
- OpenAI GPT models (gpt-3.5, gpt-4)
- Ollama local models (gemma, llama2, mistral, etc.)
Features:
- Session-based chat history (
Chat-1,Chat-2, …) - Memory: remembers past interactions in each session
- Configurable temperature & max tokens
- Hybrid LLM support (local + cloud)
A tool to query PDF documents using Retrieval-Augmented Generation (RAG). The app splits PDFs into chunks, embeds them locally using Ollama, stores in FAISS for vector similarity search, and uses Groq-hosted models for generating answers. Uses:
- Python, Streamlit
- LangChain + LangChain-community components
- Groq API for LLM inference (using a supported model)
- Ollama for embeddings
- FAISS for vector search
Features:
- Splits PDF documents into text chunks using a recursive character splitter.
- Embeds chunks via an Ollama model.
- Stores embeddings in a persistent FAISS vector store so embeddings don’t need to be recomputed on each run.
- Provides a Streamlit UI with a sidebar control to “Click for Document Embedding”, a prominent query input, and displays both the answer and similarity-matched document sections.
This project is an upgraded version of LangChain_PDF_Groq_Ollama_Embeddings.
Uses:
- Streamlit – Interactive UI
- LangChain – RAG orchestration
- Groq – Fast LLM inference (
llama-3.1-8b-instant) - HuggingFace Embeddings – Semantic search embeddings
- ChromaDB – Vector database
- pypdf – PDF text extraction
Features:
- Conversational Memory: Keeps track of session history using
RunnableWithMessageHistory. - Chat Context Awareness: Reformulates questions based on chat history (
create_history_aware_retriever). - Multiple PDF Uploads: Upload and query multiple PDFs in the same session.
- Chroma Vector Store: Uses ChromaDB instead of FAISS, optimized for conversational use.
- Streamlit UI Enhancements:
Clone the repo:
git clone https://github.com/vinit-devops/Generative-AI-Projects.git
cd Generative-AI-Projects