Multi-Dimensional Database

Vector similarity • Graph relationships • Metadata facets • AI context
Open source multi-dimensional database for AI applications.
One database, infinite dimensions.

Try Demos

⚡ Quick Start: Zero to Smart in 60 Seconds

// Brainy 1.0 - Just 9 methods. That's it. Really.
import { BrainyData, NounType, VerbType } from '@soulcraft/brainy'
const brain = new BrainyData()

// 1. init() - Start with encryption & vector search ready
await brain.init()

// 2. add() - Smart addition, understands everything
await brain.add("Claude is an AI assistant created by Anthropic")

// 3. addNoun() - Typed entities (23 types available)
const claudeId = await brain.addNoun("Claude", NounType.Product)
const anthropicId = await brain.addNoun("Anthropic", NounType.Organization)

// 4. addVerb() - Relationships (39 types available)
await brain.addVerb(claudeId, anthropicId, VerbType.CreatedBy)

// 5. search() - Multi-dimensional semantic search
const results = await brain.search("AI safety research", 10)

// 6. get() / 7. update() / 8. delete() / 9. export()
// That's it. You now know the entire API. 🎯

🎆 NEW! Talk to Your Data with Brainy Chat

// AI chat with multi-dimensional context
const response = await db.chat("What innovations are happening in automotive?")

// AI understands across all dimensions:
// • Vector: semantic meaning of "innovations"
// • Graph: relationships between companies and technologies
// • Facets: industry and sentiment filters
// • Context: your specific data and conversation history

Multi-Dimensional Intelligence: Vector dimension for semantic similarity • Graph dimension for relationship patterns • Facet dimension for metadata organization • AI dimension for natural language understanding

🚀 Runs Anywhere, Scales Everywhere

Brainy automatically adapts to your environment. Same API whether you're in a browser using OPFS, Node.js with filesystem, serverless with S3, or enterprise with cloud storage. Scale from prototype to production without changing a line of code.

🌐
Browser
OPFS Storage
💻
Node.js
Filesystem
⚡
Serverless
S3 Compatible
🏢
Enterprise
Cloud Scale

Why Developers Choose Brainy

The only multi-dimensional AI database that combines vector, graph, metadata, and AI context in one simple API

🎯

Vector Dimension

Semantic understanding through embeddings. Find similar content by meaning, not just keywords. Perfect for AI context retrieval.

🔗

Graph Dimension

Relationship traversal and connections between data points. Build knowledge graphs that understand how concepts relate.

🏷️

Facet Dimension

Metadata filtering and organization. Structure your data with categories, tags, and attributes for precise queries.

🧠

AI Dimension

Natural language queries and insights. Chat with your data and get intelligent responses powered by multi-dimensional context.

⚡

Local-First Intelligence

All dimensions work offline by default. Fast, private, and secure. Add cloud sync only when you need collaborative intelligence.

🔄

Cross-Dimensional Queries

Search across all dimensions simultaneously. Find by meaning, relationships, metadata, and context in a single powerful query.

Enterprise-Grade Performance

Production-ready with blazing fast multi-dimensional queries

2-8ms
Vector search latency
1M embeddings
1-3ms
Graph traversal
100M relations
10,000+
Queries per second
Production scale
Zero
External dependencies
Works offline

What You Can Build

🤖 AI Assistants

Perfect memory across conversations. Vector search for context, graph for relationships, metadata for organization.

🔍 Semantic Search

Beyond keywords to meaning. Combine vector similarity with metadata filters and relationship traversal.

🎯 Recommendation Engine

Smart suggestions using vectors for similarity, graphs for patterns, and facets for personalization.

🧬 Knowledge Graphs

Connect ideas and entities. Build semantic networks with automatic relationship extraction and AI insights.

👁️ Computer Vision

Visual search with context. Image embeddings plus metadata tags and relationship mapping.

🎵 Music Discovery

Audio fingerprints meet social graphs. Find similar tracks, discover through patterns, filter by genre.

🏥 Medical Diagnosis

Pattern matching across symptoms and research. Connect patient data with medical knowledge graphs.

🚨 Fraud Detection

Real-time anomaly detection. Transaction graphs, behavioral vectors, and pattern recognition.

👥 Real-time Collab

Shared intelligence across teams. Sync knowledge graphs, share context, collaborate on insights.

One Query. Three Paradigms. Infinite Possibilities.

View Full Documentation Try It Live

🚀 Start Building with CLI Commands

Now that you've seen what's possible, here's how to build it with actual working CLI commands

$ npm install @soulcraft/brainy $ npm run brainy:init ✓ Initialized local storage ✓ Created default configuration ✓ Ready to store memories! $ npm run dev > Starting application with Brainy... 
# Initialize Multi-Dimensional Database
npx brainy init --storage filesystem

# Add data across multiple dimensions (vector, graph, facets)
npx brainy add "Customer feedback about product quality" \
  --metadata '{"category":"feedback", "sentiment":"positive", "priority":"high"}'

# Multi-dimensional search across vector, graph, and facets
npx brainy search "product issues" \
  --filter '{"category":"feedback", "priority":"high"}' \
  --verbs "relates_to,caused_by" \
  --depth 2

# AI-powered chat with multi-dimensional context
npx brainy chat "What are the main product quality concerns?"

# Connect to Brain Cloud for AI memory
npx brainy connect
npx brainy cloud --status your-customer-id

🧠 AI-Powered Neural Import

Import any data and watch Brainy's AI automatically extract entities, relationships, and insights 

// AI-powered Neural Import - Brainy figures out the rest
import { NeuralImport } from '@soulcraft/brainy/cortex'

// Brainy 1.0 - Smart by default, handles everything automatically
const brain = new BrainyData()
await brain.init()

// Smart import - automatically detects types and relationships
await brain.add('./customer-feedback.csv')
// ✓ Automatically extracts 23 noun types
// ✓ Creates 39 verb relationships
// ✓ Generates embeddings for semantic search
// ✓ Universal encryption by default

// Import any data - Brainy understands it all
await brain.add({ 
  customers: loadCSV('./data.csv'),
  products: loadJSON('./products.json')
})

// Unified search across all dimensions
const insights = await brain.search("customer feedback patterns", 10)
📄

Smart File Import

CSV, JSON, YAML with AI extraction

🔮

Auto-Understanding

Entities, topics, and relationships extracted

💬

Instant Chat

Query your data conversationally

100% Free • No API keys required • Works offline • Your data stays private

☁️ Introducing Brain Cloud

Take your local Brainy database to the cloud for team collaboration and cross-device sync

Local-First, Cloud-Ready

Brainy works 100% offline and free forever. When you're ready to collaborate or sync across devices, Brain Cloud provides seamless augmentation without changing your code.

  • ✅ Free Forever: Full multi-dimensional database
  • ☁️ Cloud Sync ($19/mo): Team collaboration & device sync
  • 🏢 Enterprise ($99/mo): Dedicated infrastructure & SLA
# Enable cloud sync with one command
$ npx brainy connect

# Your local data syncs automatically
✓ Connected to Brain Cloud
✓ Syncing 1,247 memories
✓ Team workspace ready

# Same API, now with superpowers
await brain.search("insights from team", 10)
// Cloud sync happens automatically
Learn About Brain Cloud

💚 Support Open Source Brainy

Help us keep Brainy free and open source forever

Brain Cloud is our enterprise-grade AI memory infrastructure. We're committed to helping businesses build persistent AI systems that remember everything. Your sponsorship helps us advance the future of AI memory and build new features for the community.

🚀

Development

Full-time maintainers & new features

📚

Documentation

Tutorials, examples, and guides

🌍

Community

Support, events, and resources

❤️ Sponsor on GitHub 📧 Corporate Sponsorship

All sponsors get recognition in our README, priority support, and input on the roadmap.