GenAI Research & Demo Codebase

This repository contains a collection of demos, utilities, and experiments for working with Large Language Models (LLMs), chat models, embeddings, prompt engineering, and structured output parsing using LangChain, OpenAI, and related tools. It is organized for learning, research, and rapid prototyping.

Project Structure

requirements.txt
langchain_models/
    test.py
    1. LLMs/
        1_llm_demo.py
    2. ChatModels/
        1_chatmodel_openai.py
        2_chatmodel_anthropic.py
        3_chatmodel_gemini.py
        4_chatmodel_hf.py
    3. EmbeddingModel/
        1_embedding_openai_query.py
        2_embedding_openai_mul.py
        3_embedding_hf.py
        4_document_similarity.py
langchain_output_parsers/
    nodes.md
    json_schema.json
    Outputs.txt
    pydantic_demo.py
    tempCodeRunnerFile.py
    typeddict_demo.py
    with_structured_output_json.py
    with_structured_output_pydantic.py
    with_structured_output_typeddict.py
langchain_prompts/
    chat_history.txt
    chat_prompt_template.py
    chatbot.py
    message_placeholder.py
    messages.py
    notes.md
    prompt_generator.py
    prompt_ui.py
    template.json
langchain_structured_output/
    json_schema.json
    notes.md
    Outputs.txt
    pydantic_demo.py
    tempCodeRunnerFile.py
    typeddict_demo.py
    with_structured_output_json.py
    with_structured_output_pydantic.py
    with_structured_output_typeddict.py

Key Components

1. langchain_models/

• LLMs/: Demos for using various LLMs (e.g., OpenAI GPT, Anthropic, Gemini, HuggingFace).
• ChatModels/: Examples of chat-based model usage.
• EmbeddingModel/: Scripts for generating and comparing embeddings, including document similarity.

2. langchain_prompts/

• Prompt engineering utilities, templates, and UI for research paper summarization and explanation.
• Includes Streamlit-based UI (prompt_ui.py) for interactive demos.

3. langchain_output_parsers/ & langchain_structured_output/

• Demos for structured output parsing using Pydantic, TypedDict, and JSON schema.
• Example scripts for extracting structured data from LLM outputs.

Setup

1. Install dependencies:

   pip install -r requirements.txt

2. Set up environment variables:

   • Create a .env file with your API keys (OpenAI, Anthropic, etc.) as needed.

3. Run Streamlit UI:

   streamlit run langchain_prompts/prompt_ui.py

4. Run individual scripts:

   python path/to/script.py

Notable Files

• prompt_ui.py: Streamlit app for research paper summarization/explanation.
• chatbot.py: Basic chatbot implementation.
• with_structured_output_*.py: Structured output parsing demos.
• test.py: General test script for quick experiments.

Requirements

• Python 3.8+
• See requirements.txt for all dependencies.

License

This project is for educational and research purposes.