Comparison

Cognee vs Memary

On the evidence we track, Cognee leads this comparison with a composite score of 83/100. Scores are only directly comparable because these tools share a category; the full breakdown and every source is below.

Machine formatsJSONMarkdownGraphQLor send Accept: application/json
Cognee83
Memary58
Score
Vioscale score
Cognee83 / 100low · 37%
Memary58 / 100low · 27%
Pricing
Free tier
Cognee
Memary
Model
Cogneefreemium
Memaryfreemium
Price level
Cogneelow
Memaryfree
Starting price
Cognee$5
Memary
Transparent
Cognee
Memary
Integrations
Count
Cognee27
Memary3
Adoption
Dependent repos
Cognee
Memary0
Github stars
Cognee
Memary2,641
Release
Cadence days
Cognee
Memary11
History
Cognee
Memary6 items
License
Spdx
Cognee
MemaryMIT
Language
Primary
Cognee
Market

Capabilities

Feature-by-feature on the axes that matter for agent memory. “-” means undocumented, not absent.

Capabilities
Architecture model
CogneeOpen source self hosted
MemaryOpen source self hosted
Graph based entity relationship extraction
Cognee
Memary
Automatic context window rolling summarization
Cognee
Memary-
Semantic vs episodic event chronological memory
Cognee
Memary
Cross session and multi thread user persistence
Cognee
Memary
Memory conflict resolution and fact updating
Cognee
Memary-
Multi tenant user and agent isolation
Cognee
Memary
Native langchain and llamaindex memory classes
Cognee
Memary
Time decay and recency weighted retrieval
Cognee
Memary-
Zero data retention and pii redaction layer
Cognee
Memary-
SOC2 type ii
Cognee
Memary-
Mit or apache permissive oss license
Cognee
Memary-
Pricing model
Cognee-
MemaryFree open source

What each one is

The product in its own terms, so the numbers below have context.

Cognee

Leader

Cognee is a graph-based memory engine for AI agents that captures information, organizes it as interconnected knowledge, and enables agents to recall context across conversations. It runs locally, on managed cloud infrastructure, or on-premises with full control over data.

Independently observed

Memary

A memory layer for autonomous agents built on knowledge graphs. It tracks entities, relationships, and events with timestamps, supports multi-agent setups with isolated memory contexts, and integrates with LangChain and LlamaIndex to help agents evolve their knowledge through interactions.

Independently observed

Pricing

List pricing as published by each vendor, with the date we read it. Always verify at the source before you buy.

Cognee

Leader
from $5/workspace/moHybridFree tier

Free tier (1M tokens/mo). Standard from $5/workspace/mo. Enterprise custom pricing.

  • FreeFree
    • Graph-based memory
    • SDK and MCP access
    • Agentic integrations (Claude Code, Cursor, Codex)
    • Unlimited users and API calls
  • Standard$5 per workspace/month
    • Unlimited workspaces
    • Data integrations: Slack, Notion, Google Drive
    • In-app support
  • PremiumCustom pricing with enterprise support
    • Dedicated Slack channel
    • Dedicated support engineer
    • Bring-your-own cloud support
    • Support SLA
as of verify ↗

Memary

Open sourceFree tier
as of verify ↗

Platform & deployment

Where each product runs and how it can be hosted. A dash means undocumented, not unsupported.

Platforms
Web
Cognee
Memary
CLI
Cognee
Memary
Deployment
Cloud / SaaS
Cognee
Memary
Self-hosted
Cognee
Memary
On-premise
Cognee
Memary
Air-gapped
Cognee
Memary

Integrations

What each product connects to. Counts come from the vendor's own integration directory where one exists.

In common (1)
  • FalkorDB

Cognee

Leader
27 total - 26 not shared
  • Claude Code
  • Cursor
  • Codex
  • LangGraph
  • OpenClaw
  • CrewAI
  • n8n
  • Slack
  • Notion
  • Google Drive
  • Amazon Neptune
  • Weaviate
  • Qdrant
  • Pinecone
  • Redis
  • Azure AI Search
  • Milvus
  • OpenSearch
  • Valkey
  • Memgraph
  • NetworkX
  • Google Cloud Spanner
  • TuringDB
  • pgGraph
  • +2 more
Independently observed

Memary

3 total - 2 not shared
  • LangChain
  • LlamaIndex
Independently observed

Comparison generated from independently-sourced facts. Every value links to its source and retrieval date. See the method.