Superintelligence requires supermemory

supermemory is a frontier lab building memory and continual learning systems for agents.

We build the best, fastest, and most affordable way for agents to turn raw context into long-horizon knowledge for personalization and self-improvement.

Start building for freeTalk to sales
inputlearner-1memorycontext

Used in production by the world's best agents

  • Wabi
  • shapes.incshapes.inc
  • Tappz
  • Zo ComputerZo Computer
  • KeikiKeiki
  • Adapta
  • Pocket
  • Caret
  • Composio

Join 100,000+ developers building with supermemory.

The best short and long-term memory for your agent.

  • Memory that learns like the human brain

    We train a model that reasons over existing information and builds a comprehensive understanding of the entity. Knowledge builds on top of each other, and stale information gets forgotten over time.

  • Batteries included, but fully composable

    supermemory comes with retrieval, extractors for all data types, connectors to bring data in, and all the configuration you would need to tune your memory. Think of it as a framework, not a drop-in solution.

  • Permissioning and scoping built-in

    supermemory is the only memory system with deep permissioning and scoping built-in. You can query your memories, deterministically scoped just like SQL.

What people build with supermemory

  • Agent memory

    supermemory is used by popular consumer-facing agents to power memory and learning at scale for all their users.

    Especially good for

    • Personal agents
    • Knowledge agents
    • Coding agents
  • Knowledge index

    Sync data from everywhere into one index your product can search.

    SuperRAG in the docs ↗
  • Company brain

    Build your own internal company brain and knowledge tools on top of supermemory.

    company-brain on GitHub ↗
  • For personal use

    supermemory Brain

    One single memory across every agent you use.

supermemory leads across benchmarks

The combination of our model and database achieves state-of-the-art results across benchmarks.

Watch the announcementsupermemory is now the state of the art in agent memory.

supermemory is the cheapest way to do agent memory at scale.

Here is what it costs to turn 20M tokens of new text a month, about 18,000 documents, into searchable memory, once a year of it is stored.

Monthly bill after a year

  • supermemoryNothing to run

    $100
  • Graph memoryNeo4j knowledge graph, embeddings, gpt-5-mini

    $404
  • Agent-managed memoryClaude Sonnet 5.5 keeps its own memory files

    $415
  • Build it yourselfPostgres with pgvector, embeddings, gpt-5-mini

    $896

Whole monthly bill for one common configuration of each design. Model calls and embeddings are measured; database costs are estimates from list prices on November 17, 2026, one node with no standby. Engineering time is not counted.

The world's best agents use supermemory.

Moved from RAG

We just ditched RAG completely and went memory only through supermemory.
  • 40s → 12saverage response time
  • 40–50%fewer tokens
Armin DaryabegiFounder, Chatarmin

Moved from memory files

Tried almost everything — structured memory files, qmd etc. The only thing that works reliably is supermemory.
Harshil MathurFounder, Razorpay

Moved from Mem0

Mem0 was not great. Glad to have found Supermemory.
Zaid MukaddamFounder, Scira

Runs in your data center, your VPC, or on your laptop. Even air-gapped.

We build all the critical infrastructure ourselves. Which means we can deploy it wherever your data lives.

  1. 001In your data centerPlan an on-prem rollout
  2. 002In your VPCScope a VPC deployment
  3. 003On your laptopStart locally
  • AICPA SOC 2 certified
  • GDPR aligned
  • HIPAA compliant

supermemory is SOC 2 certified, HIPAA compliant and GDPR aligned. We will sign your contracts and DPA, and deploy air-gapped where the data cannot leave at all.

Questions

What is supermemory?

supermemory is a memory API for AI agents and apps. It gives any model long-term memory: it extracts memories from conversations, documents and files, keeps a profile for each user, and brings the right context back when the model needs it. It is available as an API with TypeScript and Python SDKs, as plugins for coding agents, and as an MCP server.

How is agent memory different from RAG or a vector database?

A vector database stores embeddings and returns the nearest ones; what to store, update and delete is left to you. supermemory manages the memories themselves: it extracts facts, updates them when they change, forgets what has expired and keeps a profile for each user. It also does retrieval over your documents (SuperRAG), so there is no vector database to run.

Which models, agents and frameworks does supermemory work with?

Any model. There are SDKs for TypeScript and Python and a REST API; plugins for Claude Code, Cursor, Codex, OpenCode, OpenClaw and Hermes; an MCP server for Claude, ChatGPT and other MCP clients; and integrations for the Vercel AI SDK, LangChain, LangGraph, CrewAI, the OpenAI SDK, Mastra, Pipecat, LiveKit, Zapier and n8n.

Can I self-host supermemory?

Yes. supermemory local is a free self-hosted binary for up to 10,000 stored documents; it runs on your machine and is not open source. Running supermemory in your own cloud or on a dedicated instance is available on Scale and Enterprise, and Enterprise also supports fully air-gapped deployments, where LLM inference may be the only outbound call, or none at all.

Is supermemory open source?

Partly. The main GitHub repository, including the MCP server, and the plugins are MIT, and the TypeScript and Python SDKs are Apache 2.0. The memory engine is not open source: it runs in our cloud, as the free supermemory local binary, or in your own infrastructure on Scale and Enterprise.

How much does supermemory cost?

The Free plan includes $5 of credits every month. Paid plans start at $19 a month, and usage is billed at the same rates on every plan.

Intelligence without memory is just randomness.

supermemory gives your agent long-term memory of every user. Paste one prompt into your coding agent and it wires memory into your project.

Free to start, with $5 of credits every month. Pricing · Quickstart

YOUR PRIVACY

Change or withdraw any time via Cookie settings in the footer. Read our cookie notice.