🤖 AI Agent LangGraph Chatbot

An intelligent agent-powered chatbot API built using LangGraph, Groq LLMs, and FastAPI, enhanced with real-time search using TavilySearchResults. Supports multiple large models such as LLaMA 3, Mixtral, and Gemma.

🔗 Live Repo: github.com/Srinikhil/AI-Agent-LangGraph-Chatbot

---

� Features

• 🌐 FastAPI backend for handling chat requests
• 🧠 Agent-based reasoning via LangGraph's ReAct pattern
• 🔍 Integrated web search using TavilySearchResults
• 🔄 Supports multiple LLMs via Groq:
  • llama3-70b-8192
  • distil-whisper-large-v3-en
  • gemma2-9b-it
  • mixtral-8x7b-32768
• 📡 API-first design for frontend or app integrations

---

🧠 Models Supported

Model Name	Description
llama3-70b-8192	Meta's latest high-accuracy LLM
distil-whisper-large-v3-en	Efficient speech recognition (English)
gemma2-9b-it	Fine-tuned conversational model
mixtral-8x7b-32768	Sparse Mixture-of-Experts model

---

🔐 Setup

1. Clone the repo:

git clone https://github.com/Srinikhil/AI-Agent-LangGraph-Chatbot.git
cd AI-Agent-LangGraph-Chatbot

2. Install dependencies:

pip install -r requirements.txt

3. Add API keys in app.py:

groq_api_key = 'your_groq_api_key_here'
os.environ["TAVILY_API_KEY"] = 'your_tavily_api_key_here'

4. Run

uvicorn app:app --reload
# API at: http://127.0.0.1:8000/chat

5. Run UI

streamlit run UI.py