Skip to content

🔍 RAG ChromaDB Demo Agent

The RAG ChromaDB Demo Agent showcases Retrieval-Augmented Generation capabilities with ChromaDB vector database in SuperOptiX. This demo focuses specifically on how to configure and use RAG with ChromaDB for knowledge retrieval and context-aware responses.

🎯 What This Demo Shows

This demo demonstrates:

  • 🔍 RAG Integration: How to configure RAG with ChromaDB in SuperOptiX
  • 📚 Knowledge Retrieval: Semantic search and document retrieval
  • 🧠 Context-Aware Responses: Responses based on retrieved knowledge
  • ⚙️ Playbook Configuration: How to set up RAG in agent playbooks

🚀 Setup RAG with ChromaDB

1. Install Ollama Model

Bash
# Install the Ollama model used in this demo
super model install llama3.2:8b

2. Start Ollama Server

Bash
# Start Ollama server (runs on port 11434 by default)
ollama serve

3. Pull and Run the Demo

Bash
# Pull the RAG ChromaDB demo agent
super agent pull rag_chroma_demo

# Compile the agent
super agent compile rag_chroma_demo

# Run the agent
super agent run rag_chroma_demo --goal "What is the SuperOptiX framework?"

🔧 RAG Configuration in Playbook

The RAG ChromaDB demo showcases how to configure RAG in the agent playbook:

Language Model Configuration

YAML
language_model:
  location: local
  provider: ollama
  model: llama3.2:8b
  api_base: http://localhost:11434
  temperature: 0.7
  max_tokens: 2048

RAG Configuration

YAML
rag:
  enabled: true
  retriever_type: chroma
  config:
    top_k: 5
    chunk_size: 512
    chunk_overlap: 50
    similarity_threshold: 0.7
  vector_store:
    embedding_model: sentence-transformers/all-MiniLM-L6-v2
    collection_name: rag_demo_knowledge
    persist_directory: ./data/chromadb

Key RAG Configuration Points:

  • ✅ enabled: true: Enables RAG functionality
  • 🗄️ retriever_type: chroma: Uses ChromaDB as vector database
  • 🔝 top_k: 5: Retrieves top 5 most similar documents
  • 📄 chunk_size: 512: Document chunk size for processing
  • 🔗 chunk_overlap: 50: Overlap between chunks for context
  • 🎯 similarity_threshold: 0.7: Minimum similarity score for retrieval
  • 🧠 embedding_model: Sentence transformers for embeddings
  • 📁 collection_name: ChromaDB collection name
  • 💾 persist_directory: Local storage directory

🔍 RAG: Your AI's Memory Bank

Retrieval-Augmented Generation (RAG) gives your AI agent the ability to access and use specific knowledge. Think of it as giving your AI a personal library:

  • 🧠 Semantic Understanding: Finds relevant information based on meaning, not just keywords
  • 📚 Knowledge Base: Access to your own documents, databases, and information sources
  • 🎯 Precise Answers: Generates responses based on actual retrieved content
  • 📖 Source Citations: Always provides references to where information came from
  • 🔄 Real-time Updates: Can access the latest information as it becomes available

🔧 Customizing RAG Configuration

Adjust Retrieval Settings

Edit agents/rag_chroma_demo/playbook/rag_chroma_demo_playbook.yaml:

YAML
rag:
  config:
    top_k: 10  # Retrieve more documents
    chunk_size: 1024  # Larger chunks
    similarity_threshold: 0.8  # Higher similarity threshold

Change Vector Database

YAML
rag:
  retriever_type: lancedb  # Use LanceDB instead
  vector_store:
    table_name: lancedb_demo_table
    database_path: ./data/lancedb

Use Different Embedding Model

YAML
rag:
  vector_store:
    embedding_model: sentence-transformers/all-mpnet-base-v2  # Different embeddings

🚨 Troubleshooting RAG

Common Issues

  1. Ollama Server Not Running

    Bash
    # Check if Ollama server is running
curl http://localhost:11434/api/tags

# Start Ollama server
ollama serve

  2. ChromaDB Issues

    Bash
    # Check RAG configuration
super agent inspect rag_chroma_demo

# Clear ChromaDB data if needed
rm -rf ./data/chromadb

  3. RAG Not Working

    Bash
    # Check RAG configuration
super agent inspect rag_chroma_demo

# Verify knowledge base is populated

Getting Help

Bash
# Check agent status
super agent inspect rag_chroma_demo

# View agent logs
super agent logs rag_chroma_demo

# Get RAG help
super agent --help

🎉 Next Steps

After exploring the RAG ChromaDB demo:

  1. Try Other RAG Backends: Explore RAG LanceDB for production use
  2. Customize: Modify the playbook for your specific knowledge base
  3. Build Your Own: Use this as a template for your custom RAG agent

Ready to explore knowledge retrieval? Start with the RAG ChromaDB demo! 🚀 Ready to explore knowledge retrieval? Start with the RAG ChromaDB demo! 🚀