Docs/Getting Started

Recall Documentation

Zero-configuration memory for AI. Works instantly with embedded Redis, optional cloud sync.

What's New in v1.3.0

  • AI Intelligence - Real vector embeddings, entity extraction, knowledge graphs (NEW!)
  • Semantic Search - Find memories by meaning, not just keywords
  • Local-First Mode - No setup required, embedded Redis starts automatically
  • 100% TypeScript - Full type safety and IntelliSense support
  • Zero Config - Just run npx r3 and it works

Start building

Quickstart →

Get started in 30 seconds - zero configuration

AI Intelligence

NEW

Semantic search, entities, knowledge graphs

API Reference →

Complete reference for all MCP tools

How Recall works

Recall is an MCP server that gives AI applications persistent memory. By default, it uses an embedded Redis server for local storage - no setup needed. Optionally add cloud sync with Mem0.

JSON
1// ~/.claude/claude_desktop_config.json
2{
3  "mcpServers": {
4    "recall": {
5      "command": "npx",
6      "args": ["r3"]
7      // No environment variables needed!
8    }
9  }
10}

Once configured, Claude can remember context across conversations:

TEXT
1You: Remember that I prefer TypeScript over JavaScript
2Claude: I'll remember your preference for TypeScript.
3

4[New conversation]
5You: What language should I use for this project?
6Claude: Based on your preferences, I'd recommend TypeScript.

Architecture

Claude Desktop
MCP Protocol
r3 Server
Redis(L1 Cache)
Mem0 Cloud(L2 Storage)
<5ms
Cache Hit
~200ms
Cache Miss
~10ms
First Store

Core concepts

Memories

Each memory contains:

  • Content: The information to remember
  • User ID: Who it belongs to
  • Priority: How important it is (affects caching)
  • Metadata: Additional context

Caching layers

  1. L1 (Redis): Hot data, 24-hour TTL
  2. L2 (Cloud): Permanent storage with vector search

MCP tools

Recall provides these tools to Claude:

  • add_memory: Store new information
  • search_memory: Find relevant context
  • get_all_memories: List all memories
  • delete_memory: Remove specific memory
  • cache_stats: Monitor performance

When to use Recall

Good for:

  • User preferences and settings
  • Conversation history
  • Facts about users or projects
  • Context that should persist

Not for:

  • Sensitive data (use encryption)
  • Large files (use object storage)
  • Real-time data (use streaming)

Pricing

Recall itself is free and open source. You'll need:

  • Redis: Free locally, ~$15/mo for Redis Cloud
  • Mem0: Free tier includes 10,000 memories

Get help

Quick Start