Feature Tour

Intelligence, Decentralized.

From local vectors to hierarchical knowledge graphs. See what makes SuperLocalMemory the most advanced local memory system.

NEW

Adaptive Learning (v2.7)

Learns your preferences and workflows locally using machine learning. Surfaces the right memories for the right project context.

View Docs

Privacy First

GDPR compliant by design. Your behavioral data is isolated in a dedicated local database and never leaves your machine.

View Docs

Hybrid Search

Combines Semantic (Vector), Full-Text, and Graph traversal. Measured 10.6ms median search at 100 memories, sub-200ms at 1,000.

View Docs

Measured Performance

10.6ms median search. 220 writes/sec peak. Zero lock errors across 10 concurrent agents. 13.6 MB for 10,000 memories. Real benchmarks, not estimates.

View Docs

Universal Integration

Works with Cursor, Windsurf, VS Code, Claude Desktop, and CLI simultaneously.

View Docs

10-Layer Universal Architecture

The only memory system built with a complete vertical stack, from raw SQLite storage up to real-time visualization and adaptive learning.

Explore Full Architecture

architecture-stack.sh

Layer 9: VISUALIZATION (v2.2.0)

Layer 8: HYBRID SEARCH (v2.2.0)

Layer 7: UNIVERSAL ACCESS

Layer 6: MCP INTEGRATION

Layer 5.5: ADAPTIVE LEARNING (v2.7)

Layer 5: SKILLS LAYER

Layer 4: PATTERN LEARNING + MACLA

Layer 3: KNOWLEDGE GRAPH + HIERARCHICAL

Layer 2: HIERARCHICAL INDEX

Layer 1: RAW STORAGE (SQLite + FTS5)

Layer 1: Cross-Project

Transferable Tech Preferences

Layer 2: Context

Project Signal Detection

Layer 3: Workflow

Workflow Sequence Mining

Your AI Learns You

Most memory systems just store text. We store patterns. SuperLocalMemory observes which memories you actually use and re-ranks future results based on your unique workflow.

3-Layer Local ML Model
ML-Powered Adaptive Ranking
Zero Telemetry (Training is Local)

COMMON QUESTIONS

Frequently Asked Questions

What is a knowledge graph in SuperLocalMemory?

The knowledge graph automatically extracts entities and relationships from your memories, connecting related concepts. This enables graph-based search that finds relevant memories even when keywords don't match.

What is memory lifecycle management?

Memory lifecycle automatically ages memories through states (Active, Warm, Cold, Archived) based on usage. This keeps your system fast by prioritizing recent, relevant memories while preserving old ones.

What is behavioral learning?

Behavioral learning tracks what happens after you use a memory. If a recalled memory leads to a successful action, similar memories get boosted in future searches. No LLM inference required.

Ready to Try It?

Give your AI tools persistent memory in minutes.

Download Desktop App CLI Install Guide