⚡ Advanced LLM-Powered Application using LangChain

This project showcases an advanced LLM-based application using LangChain to integrate multiple language models and embedding providers including OpenAI, Claude (Anthropic), Google Gemini AI, and Hugging Face. It supports dynamic prompt handling, multi-model querying, and flexible output generation based on user input or business logic. Embedding models are used to power semantic search and retrieval-based tasks.

---

🚀 Key Features

• 🔗 Multi-LLM integration: OpenAI (GPT-4), Claude, Google Gemini, and Hugging Face (local and API-based)

• 🧠 Chat-based workflows using ChatOpenAI, ChatGoogle, ChatAnthropic, and ChatHuggingFace

• 🧩 Modular architecture using LangChain's prompt templates, chains, and tools

• 📡 Real-time data fetching, user input processing, and SQL query generation via LLMs

• 💬 Natural language to SQL conversion using integrated language models

• 🧠 Embeddings integration:

  • OpenAI embeddings (OpenAIEmbeddings)
  • Hugging Face embeddings (HuggingFaceEmbeddings) used for semantic document similarity
  • Supports both local and hosted models

• 🔐 Secure API key handling via .env for all providers

---

🛠 Built Using

• Python
• LangChain
• OpenAI API
• Claude (Anthropic) API
• Google Gemini API
• Hugging Face Inference API & Local Models

---

🧰 Tech Stack

• Python
• LangChain
• OpenAI GPT
• Anthropic Claude
• Google Gemini
• Hugging Face (Local & API models)
• Hugging Face & OpenAI Embeddings