mem-dog — The Private AI Memory Platform

Apache 2.0 · Free Forever · Self-Hosted

The Private AI Memory Platform

Ingest from 300+ apps. Enrich with 42 AI agents. Query with 5 search modes. Self-hosted and free.

bash
$docker compose up
# UI: localhost:3000  | API: localhost:8080  | Neo4j: localhost:7474

Run locally in one command. No account needed.

300+
App Integrations
42
AI Agents
5
Search Modes
70+
API Endpoints

Why mem-dog?

AI should be private. Not as a feature.

Every time you ask an AI a question, your data travels to a distant data center. mem-dog ends that trade-off.

Private by Design

Data never leaves your network. Self-hosted with local-first models via Ollama — runs fully offline. No telemetry, no phone-home. Apache 2.0 licensed.

Speed Through Smarter Architecture

A 6-layer classification pipeline matches each piece of data to the smallest model that can handle it. Simple text gets a 4B model. Complex docs get a 27B one. Most queries finish in milliseconds.

Cost Efficiency Through Local Inference

Cloud AI bills grow linearly with usage. With local Ollama models, the marginal cost of a query is the electricity to run it. No per-seat licensing. No surprise invoices.

Intelligence Through Better Data Processing

42 agents work in concert — classifying, extracting entities, generating embeddings, building knowledge graphs, and creating AI-powered viewpoints — all locally. Gets smarter the more you use it.

mem-dog = mem0 (memory) + Zep (knowledge graph) + Nango (integrations) + Dify (AI workflows) — in one platform.

How it Works

Data flows from any source through a real-time AI pipeline

Into a private, queryable data layer — all automatic, all configurable.

01/06

Data Sources

300+ apps

  • WhatsApp
  • Slack
  • Email
  • Telegram
  • Discord
  • Nango OAuth
  • Files
  • Webhooks (whk_*)
02/06

Ingest Pipeline

Normalize & route

  • Per-User Webhooks
  • Universal Envelope
  • NATS Streaming
  • Nango Credentials
03/06

AI Processing

42 agents

  • 6-Layer Classification
  • Smart Model Routing
  • Viewpoint Generation
  • Embedding Creation
04/06

Memory Layer

10 memory types

  • Versioned Storage
  • pgvector Embeddings
  • Per-User Scoping
  • Encrypted Credentials
05/06

Query & Search

Semantic RAG

  • Natural Language
  • Cosine Similarity
  • Inline Citations
  • 5 Search Modes
06/06

DigiMe Agent

15+ Channels

  • WhatsApp · Slack · Telegram
  • Conversational RAG
  • Ingest via Chat
  • MCP Server

Features

Flexible. Powerful. Complete.

Everything you need to build a private AI memory platform for your data.

300+ App Integrations

Connect Slack, Gmail, GitHub, WhatsApp, Telegram, Discord, HubSpot, Salesforce, and 300+ more via Nango. OAuth2 with automatic token refresh and AES-256-GCM credential encryption.

42-Agent AI Pipeline

Every piece of data is automatically classified, analyzed, summarized, and embedded by specialized agents. 60+ data types: PDFs, images, video, audio, code, medical records, geospatial data, IoT telemetry.

5 Search Modes

Vector (pgvector cosine), full-text (BM25), hybrid (vector + BM25 via RRF), graph (Graphiti BFS on Neo4j), and full (all signals merged). Plus 4 reranking strategies.

Temporal Knowledge Graph

Dual-layer: Postgres + Graphiti/Neo4j. Facts have valid_at/invalid_at timestamps for point-in-time retrieval. "Who was CEO of Acme in 2024?" just works.

RAG Chat with Citations

Conversational answers grounded in your data. Every claim is backed by numbered [1][2] citations linking back to source documents. Works with all 5 search modes.

10 Memory Types

Timeline, session, conversation, user, organizational, factual, episodic, semantic, custom, tracing. Default TTLs per type, ACLs (private/shared/public/restricted).

Smart Model Routing

5 model tiers: small (4b) to omni. Per-agent-type assignment with fallback chains (Ollama local → Ollama Cloud → Gemini). Model Garden supports 10+ AI providers.

Per-User Webhooks

Each user gets unique webhook endpoints (whk_<ulid>) with HMAC-SHA256 verification, event logging, and stats. 25+ channel adapters normalize payloads into a universal format.

Multi-Language SDKs

Python, TypeScript, Go, Rust, Ruby. ~120 methods each, plus a 7-method simple facade. Agent adapters for LangChain, CrewAI, and OpenAI function calling. MCP server for Claude Desktop.

Self-Hosted & Free

Run on Docker Compose (local dev), GKE, or any Kubernetes cluster. No vendor lock-in, no usage fees. Apache 2.0 license. Your data never leaves your infrastructure.

System Architecture

Built for production

Real-time processing, tenant isolation, and complete observability. Run locally or on Google Cloud.

300+ Apps
Slack, WhatsApp, Email...
Web UI (Next.js)
Upload, Chat, Playground
MCP Clients
Claude Desktop, Cursor
Webhook Gateway
Normalize → Route
API (FastAPI)
Ingest, Query, Manage
NATS Pipeline
42 AI Agents
Supabase
Postgres + pgvector
Neo4j
Graphiti KG
Nango
OAuth Tokens
Search + RAG
Vector · FTS · Graph · Hybrid
ComponentStackDeployment
APIPython 3.12, FastAPIGKE
UINext.js 14, TypeScriptCloud Run
MCP ServerPython 3.12, FastMCP, SSEGKE
Webhook PipelinePython 3.12, NATS, 42 agentsGKE
Webhook GatewayPython 3.12, FastAPI, LiteLLMGKE
Neo4jNeo4j 5.26 Community + GraphitiGKE
NangoOAuth, token refresh, encryptionGKE
SDKsPython, TypeScript, Go, Rust, Rubynpm/pip/cargo

Search Deep-Dive

5 Search Modes. 4 Rerankers.

Not just vector search. Every signal, every strategy, every query type — covered.

Search Modes

vector
pgvector (cosine similarity)
Semantic meaning, concept matching
fts
BM25 keyword search
Exact terms, names, codes
hybrid
Vector + BM25 via RRFrecommended
General-purpose (recommended default)
graph
Graphiti BFS on Neo4j
Relationship traversal, temporal facts
full
All signals merged
Maximum recall, complex questions

Rerankers

noneReturn results as-is
rrfReciprocal Rank Fusion — merge multiple rankings
mmrMaximal Marginal Relevance — boost diversity
cross-encoderLLM-scored relevance — highest quality
from mem_dog_client import MemDog
m = MemDog("http://localhost:8080")

# Hybrid search with MMR reranking
results = m.search(
    "project timeline decisions",
    mode="hybrid",
    reranker="mmr",
    limit=10
)

# RAG Chat with citations
answer = m.chat(
    "What decisions were made?",
    search_mode="hybrid",
    reranker="cross-encoder"
)
print(answer.text)
# [1] Decision about roadmap...
# [2] Q3 planning notes...

Comparison

How mem-dog compares

Each tool does one thing well. mem-dog combines all of them — integrations, AI processing, memory, search, and a knowledge graph — into a single self-hosted system.

← Scroll to compare →

FeatureMem-DogDify.aiMem0ZepLangMem
Core Architecture
Runs fully on-premise / at home
Self-hosted AI models (Ollama)
Near-zero marginal cost per query
Data never leaves your network
Single docker compose up
Data & Integrations
300+ app integrations (Nango)
Per-user webhook endpoints
Multi-channel ingest (WhatsApp, Slack)
File/image/video/audio ingest
Versioned storage with full history
AI & Intelligence
42 specialized AI agents
5 model tiers with smart routing
Knowledge graph (Neo4j/Graphiti)
Temporal fact tracking
RAG chat with inline citations
Memory & Search
10 memory types
5 search modes
Memory compression (LLM summarization)
Per-user scoping & ACLs
Developer Experience
Python SDK with LangChain/CrewAI adapters
Multi-language SDKs (TS, Go, Rust, Ruby)
Built-in conversational agent (DigiMe)
Interactive playground in UI

mem-dog = Nango connects your apps. Dify builds AI workflows. Mem0 adds memory. Zep tracks facts. mem-dog does all of it — on hardware you own, at a fraction of the cost.

Use Cases

Built for every use case

From personal knowledge management to enterprise compliance — one platform, infinite possibilities.

Personal Knowledge Base

Capture from WhatsApp, email, Slack. Semantic search across everything. Never lose a conversation or idea.

Team Memory

Shared org memory across channels. Auto-classify meetings, decisions, action items. Queryable by anyone.

Customer Intelligence

Ingest support tickets, CRM, chat logs. AI extracts sentiment and trends. 300+ app connections via Nango.

Research & Analysis

Ingest PDFs, papers, web pages, datasets. AI viewpoints and summaries. Semantic connections across sources.

Compliance & Audit

Every mutation versioned. OpenTelemetry tracing. Per-item ACLs. Immutable audit trail.

IoT & Sensor Data

GPS, biometric, weather, industrial sensors. Specialized agents for time-series and geospatial data.

Legal & Contract Intelligence

Ingest contracts, NDAs, legal briefs. AI extracts clauses, obligations, deadlines. Temporal graph tracks amendments.

Healthcare & Clinical Notes

Process medical records, imaging reports, lab results. DICOM-aware agents. Knowledge graph links patients, conditions, treatments.

Education & Training

Ingest lectures, textbooks, course materials. AI generates study guides and flashcards. Students query their learning history.

Sales Enablement

Connect Salesforce, HubSpot, email. AI summarizes deal history, extracts action items. Search across all customer touchpoints.

Media Monitoring

RSS feeds, social media, news APIs. Real-time sentiment analysis. Track brand mentions across channels with temporal trends.

Meeting Intelligence

Zoom, Teams, Google Meet recordings. AI transcribes, extracts decisions, action items. Searchable meeting memory across your org.

Developer

Built for developers

SDKs in 5 languages, agent framework adapters, and a full REST API. Get started in minutes.

Quick Start

bash
# 1. Clone & start
$git clone https://github.com/BuildGeekAI/mem-dog.git
$cd mem-dog && docker compose up
# UI: localhost:3000  | API: localhost:8080
# Neo4j: localhost:7474  | 3 Ollama tiers

Developer Highlights

  • SDKs: Python, TypeScript, Go, Rust, Ruby — full API coverage
  • Adapters: LangChain, CrewAI, OpenAI Agents SDK
  • Graph Memory: Temporal knowledge graph with Neo4j + Graphiti
  • Memory Compression: LLM-powered summarization for long-term recall
  • MCP Server: 8 tools for Claude Desktop, Cursor, and more
from mem_dog_client import MemDog
m = MemDog("http://localhost:8080")

# Add data
m.add("Project kickoff notes...", tags=["projects"])

# Semantic search
results = m.search("project timeline", mode="hybrid")

# RAG chat
answer = m.chat("What are the key milestones?")

# Knowledge graph
entities = m.entities(query="people involved")
related = m.related(entity_id="ent_01ABC...")

Also available in Go, Rust, and Ruby. Full API docs at /docs.

FAQ

Frequently Asked Questions

Ready to try mem-dog?

Get running in under 2 minutes. No account needed.

$ git clone https://github.com/BuildGeekAI/mem-dog && cd mem-dog && docker compose up

Enterprise inquiries: [email protected]