requirements.txt

# LangChain Framework and Extensions
# Core LangChain library for building LLM applications
langchain==1.2.6
# Community integrations for LangChain (vector stores, document loaders, etc.)
langchain-community==0.4.1
# OpenAI integration for LangChain
langchain-openai==1.1.7
# HuggingFace model integration for LangChain
langchain_huggingface==1.2.0
# Text splitting utilities for document processing
langchain-text-splitters==1.1.0

# Document Processing
# PDF file reading and manipulation
PyPDF2==3.0.1

# Configuration and Environment
# Load environment variables from .env files
python-dotenv==1.2.1

# Web Application Framework
# Interactive web application framework for data apps
streamlit

# AI/ML APIs and Libraries
# OpenAI API client for GPT models
openai==1.109.1
# Facebook AI Similarity Search for efficient vector storage (CPU version)
faiss-cpu==1.13.2
# Declarative visualization library for Streamlit
altair==6.0.0
# OpenAI's tokenizer for counting tokens
tiktoken==0.12.0
# HuggingFace model hub client
huggingface-hub==0.36.0
# Instruction-based embedding models
InstructorEmbedding==1.0.1
# Sentence embeddings and semantic similarity
sentence_transformers==5.2.0

# Code Quality and Development Tools
# Python code formatter
black==26.1.0
# Python linter for code quality
pylint==4.0.4
# File system event monitoring
watchdog==6.0.0

# RAG Evaluation and Quality Assurance
# RAG (Retrieval-Augmented Generation) evaluation framework
ragas==0.4.3
# AI output validation and guardrails
guardrails-ai==0.7.2
# Data manipulation and analysis
pandas==2.3.3
# HuggingFace datasets library
datasets==4.5.0
# Fast fuzzy string matching
rapidfuzz

# API Development
# Modern web framework for building APIs
fastapi==0.128.0
# ASGI server with standard extras (websockets, etc.)
uvicorn[standard]
# Multipart form data parsing for file uploads
python-multipart==0.0.21
# Data validation using Python type annotations
pydantic==2.12.5
# HTTP client for making API requests
gunicorn