Memory
AI Memory Overview
Last updated March 3, 2026
Vector storage for RAG, conversation history, and semantic search.
AI Memory provides vector storage for building RAG applications, conversation history, and semantic search.
Key Features
- Automatic Embeddings - Content is embedded automatically when stored
- Semantic Search - Find relevant content using natural language
- Namespaces - Organize memories by project, user, or topic
- Metadata Filtering - Filter search results by custom metadata
- RAG Helper - Built-in retrieval-augmented generation
Quick Start
import { Cencori } from 'cencori';
const cencori = new Cencori({ apiKey: 'csk_...' });
// Store a memory
await cencori.memory.store({
namespace: 'docs',
content: 'Refund policy allows returns within 30 days',
metadata: { category: 'policy', version: '2024' }
});
// Search memories
const results = await cencori.memory.search({
namespace: 'docs',
query: 'what is the refund policy?',
limit: 5
});
console.log(results[0].content);Use Cases
Knowledge Base RAG
// Store documentation
await cencori.memory.store({
namespace: 'knowledge-base',
content: 'Product documentation content...',
metadata: { docId: 'doc-123', category: 'product' }
});
// Query with context
const response = await cencori.ai.rag({
model: 'gpt-4o',
messages: [{ role: 'user', content: 'How do I reset my password?' }],
namespace: 'knowledge-base'
});Conversation History
// Store conversation
await cencori.memory.store({
namespace: `user-${userId}`,
content: 'User asked about pricing...',
metadata: { sessionId, timestamp: Date.now() }
});
// Recall relevant context
const context = await cencori.memory.search({
namespace: `user-${userId}`,
query: 'What did we discuss about pricing?'
});Documentation
- Namespaces - Organize memories
- Store & Search - Core operations
- RAG - Retrieval-augmented generation