Skip to main content
Supermemory is context infrastructure for AI agents. We’re one of the leading memory providers, with components to go beyond memory and configure it to be perfect for every usecase. It provides all the building blocks — Memory, Retrieval, Profiles, Connectors, Extractors, Evals, observability, and more.
With supermemory, developers can provide perfect recall about their users to build AI agents that are more intelligent, more personalized, and more consistent. It is the state of the art across multiple different benchmarks, like LongMemEval and LoCoMo. It’s also the best in a lot of independantly run benchmarks, like the SWEContext bench.

​
How does it work? (at a glance)

  • You send Supermemory raw data in any format - text, files, and chats, or connect it to the data sources
  • Supermemory intelligently indexes them using our user understanding model and builds a semantic understanding graph on top of an entity (e.g., a user, a document, a project, an organization). We call these entities containerTag
  • This knowledge is now traversed by the agent, and an automatic profile is built for it. The agent may now use it for memory operations or for retrieval.

​
Why add memory to your agent?

Without memory, every session starts from zero. The model cannot know what the user preferred last week, which project they are on, or that a fact has changed since yesterday. Memory gives an agent durable understanding of people and entities over time — preferences, decisions, relationships, corrections. Retrieval (RAG) grounds answers in documents and knowledge bases. You usually want both. By adding memory to your agent, you can:
  • Personalize — remember preferences, roles, and history across sessions without stuffing the full chat log into every prompt
  • Stay correct as facts change — “I love Adidas” then “switching to Puma” should not leave both preferences equally true
  • Ground answers — pull the right policy, ticket, or doc when the question needs source material
  • Ship multi-tenant products — isolate each user or workspace so one customer’s memory never leaks into another’s
You can think of memory as the always-on context a skilled teammate would carry — not a search box over raw logs. For the full category argument (vs DIY vectors, thin memory wrappers, pure RAG), see Comparison and Memory vs RAG.

​
Why Supermemory?

  • State of the art on long-horizon memory — #1 on LongMemEval, LoCoMo, and ConvoMem, plus independent benches like SWEContext
  • Memory is a graph, not a blob store — facts update, connect, and forget in real time; not nearest-neighbor chunks alone
  • User profiles built in — static + dynamic context the agent should always know, ~ready for the prompt
  • Memory + SuperRAG in one engine — personalize and ground on the same containerTag / context pool
  • Every door, one store — API, MCP, plugins, SMFS, connectors, and Company Brain share the same memories
  • Multimodal by default — text, chats, PDFs, images, video, code via extractors and connectors
  • Run it your way — managed cloud or self-host as a single binary (including offline)
memory graph
Memory, profiles, and SuperRAG share the same context pool when you use the same isolation (containerTag). Mix and match for your product! A container can be anything - a user, a project, team, organization, etc.

​
Next steps

Quickstart

Make your first API call in minutes

How it Works

Understand the knowledge graph architecture

Comparison

vs DIY vectors, thin memory layers, pure RAG

Self-host it

One binary, zero config, fully offline

Billing & plans

Credits, SM tokens, and how usage works

Security & compliance

SOC 2, GDPR, HIPAA BAA, encryption