Skip to main content

​
Vector Store API

Configure the vector storage backend for the knowledge base.

​
Flag

--vector-store <name>

​
Available Stores

NameDescriptionRequirements
memoryIn-memory storeNone (built-in)
chromaChromaDBpip install chromadb
pineconePineconepip install pinecone-client
qdrantQdrantpip install qdrant-client
weaviateWeaviatepip install weaviate-client

​
Usage

​
With Query Command

# Use in-memory store
praisonai knowledge query "What is Python?" --vector-store memory

# Use ChromaDB (default)
praisonai knowledge query "What is Python?" --vector-store chroma

# Use Pinecone
praisonai knowledge query "What is Python?" --vector-store pinecone

# Use Qdrant
praisonai knowledge query "What is Python?" --vector-store qdrant

​
With Add Command

# Add to specific store
praisonai knowledge add document.pdf --vector-store chroma

​
Environment Variables

Some stores require API keys: 
# Pinecone
export PINECONE_API_KEY=your-key
export PINECONE_ENVIRONMENT=us-east-1

# Qdrant Cloud
export QDRANT_API_KEY=your-key
export QDRANT_URL=https://your-cluster.qdrant.io

​
Examples

​
In-Memory Store (Testing)

# Quick testing without persistence
praisonai knowledge add test.txt --vector-store memory
praisonai knowledge query "test query" --vector-store memory

​
ChromaDB (Local Persistence)

# Default - persists to .praison/knowledge/
praisonai knowledge add docs/ --vector-store chroma
praisonai knowledge query "How to authenticate?"

​
Cloud Stores

# Pinecone (requires API key)
export PINECONE_API_KEY=pk-xxx
praisonai knowledge query "query" --vector-store pinecone

# Qdrant Cloud
export QDRANT_API_KEY=xxx
praisonai knowledge query "query" --vector-store qdrant

​
See Also