mirror of
https://github.com/supermemoryai/supermemory.git
synced 2026-10-07 02:58:11 +00:00
640eeaa8e8
Co-authored-by: Dhravya Shah <dhravya@supermemory.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
175 lines
12 KiB
Text
175 lines
12 KiB
Text
---
|
|
title: "What is Supermemory?"
|
|
description: "Supermemory is the long term and short term context and memory infrastructure for agents."
|
|
sidebarTitle: "What is Supermemory?"
|
|
icon: "/icons/hugeicons/book-open-01.svg"
|
|
---
|
|
|
|
export const ArtTile = ({ title, href, art, pos = "center", flip, children }) => {
|
|
return (
|
|
<a className="sm-card sm-art-card group" href={href}>
|
|
<div className="sm-card-media sm-art-media">
|
|
<img src={art} alt="" style={{ objectPosition: pos, "--flip": flip ? -1 : 1 }} />
|
|
<div className="sm-art-wash" />
|
|
<div className="sm-scene" aria-hidden="true">{children}</div>
|
|
</div>
|
|
<div className="sm-card-body">
|
|
<div className="sm-card-title">
|
|
<span>{title}</span>
|
|
<svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" strokeWidth="2" strokeLinecap="round" strokeLinejoin="round" aria-hidden="true"><path d="M5 12h14M13 6l6 6-6 6" /></svg>
|
|
</div>
|
|
</div>
|
|
</a>
|
|
)
|
|
}
|
|
|
|
export const At = ({ x, y, w, className, children }) => (
|
|
<div className={className} style={{ left: x, top: y, width: w }}>{children}</div>
|
|
)
|
|
|
|
export const Tile = ({ x, y, size = 36, src, icon, logo, on, faded, badge }) => (
|
|
<div className={"m m-tile" + (src ? (logo ? " is-logo" : " is-app") : "") + (on ? " is-on" : "") + (faded ? " is-faded" : "")} style={{ left: x, top: y, width: size, height: size, padding: logo ? Math.round(size * 0.22) : 0 }}>
|
|
{src ? <img src={src} alt="" /> : <Icon icon={icon} size={Math.round(size * 0.45)} />}
|
|
{badge ? <span className="m-badge"><Icon icon="/icons/hugeicons/tick-02.svg" size={9} /></span> : null}
|
|
</div>
|
|
)
|
|
|
|
export const Lines = ({ d }) => (
|
|
<svg className="m-lines" viewBox="0 0 264 152" aria-hidden="true">
|
|
{d.map((path) => <path key={path} d={path} />)}
|
|
</svg>
|
|
)
|
|
|
|
Supermemory is **context infrastructure for AI agents**. It gives your agent memory, retrieval and user profiles through one API, and you can configure each part for your use case.
|
|
|
|
These are the building blocks it ships with:
|
|
|
|
<div className="not-prose sm-art-grid">
|
|
<ArtTile title="Memory & continual learning" href="/concepts/graph-memory" art="/images/art/bg-branch.jpg" pos="center">
|
|
<At x={48} y={8} w={168} className="m m-card m-row is-faded is-back"><span className="m-ic"><Icon icon="/icons/hugeicons/source-code.svg" size={12} /></span><s>Prefers Python</s></At>
|
|
<At x={32} y={38} w={200} className="m m-card m-row is-front"><span className="m-ic is-on"><Icon icon="/icons/hugeicons/refresh.svg" size={12} /></span><span className="m-grow">Uses TypeScript</span><span className="m-tag">Updates</span></At>
|
|
<Lines d={["M132 82 L 132 104"]} />
|
|
<At x={76} y={104} className="m m-chip"><Icon icon="/icons/hugeicons/plus-sign.svg" size={11} />Bun for scripts</At>
|
|
</ArtTile>
|
|
<ArtTile title="SuperRAG (retrieval)" href="/concepts/super-rag" art="/images/art/bg-meadow.jpg" pos="center">
|
|
<At x={24} y={14} w={216} className="m m-card m-row m-search"><Icon icon="/icons/hugeicons/search-01.svg" size={13} /><span className="m-grow">refund policy for annual plans</span></At>
|
|
<At x={56} y={94} w={152} className="m m-card m-ghost is-faded" />
|
|
<At x={44} y={82} w={176} className="m m-card m-ghost is-faded" />
|
|
<At x={32} y={66} w={200} className="m m-card m-row is-front"><span className="m-ic is-on"><Icon icon="/icons/hugeicons/file-02.svg" size={12} /></span><span className="m-grow"><span className="m-skel" style={{ width: "88%" }} /><span className="m-skel" style={{ width: "56%" }} /></span><span className="m-score">0.92</span></At>
|
|
</ArtTile>
|
|
<ArtTile title="Filesystems" href="/smfs/overview" art="/images/art/bg-stream.jpg" pos="center">
|
|
<At x={21} y={38} className="m-folder"><span className="m-folder-front"><img src="/images/art/mark-white.svg" alt="" /></span></At>
|
|
<Lines d={["M111 74 L 131 74"]} />
|
|
<At x={135} y={26} w={108} className="m m-card m-files">
|
|
<div className="m-file is-on"><Icon icon="/icons/hugeicons/file-02.svg" size={12} /><span>profile.md</span></div>
|
|
<div className="m-file"><Icon icon="/icons/hugeicons/file-01.svg" size={12} /><span>notes.md</span></div>
|
|
<div className="m-file"><Icon icon="/icons/hugeicons/folder-01.svg" size={12} /><span>meetings</span></div>
|
|
</At>
|
|
</ArtTile>
|
|
<ArtTile title="Profiles" href="/concepts/user-profiles" art="/images/art/bg-grass.jpg" pos="center">
|
|
<At x={24} y={30} w={216} className="m m-card m-profile">
|
|
<div className="m-profile-head">
|
|
<span className="m-avatar"><img src="/images/art/maya.jpg" alt="" /></span>
|
|
<span className="m-grow"><b>Maya Kapoor</b><span>Payments engineer</span></span>
|
|
</div>
|
|
<div className="m-fact"><span className="m-ic"><Icon icon="/icons/hugeicons/location-01.svg" size={11} /></span><span className="m-grow">Based in Berlin</span><span className="m-fact-kind">Static</span></div>
|
|
<div className="m-fact"><span className="m-ic is-on"><Icon icon="/icons/hugeicons/flash.svg" size={11} /></span><span className="m-grow">Fixing webhooks</span><span className="m-tag">Dynamic</span></div>
|
|
</At>
|
|
</ArtTile>
|
|
<ArtTile title="Connectors" href="/connectors/overview" art="/images/art/bg-lakeshore.jpg" pos="center">
|
|
<Lines d={["M78 30 C 130 30, 140 76, 176 76", "M78 76 L 176 76", "M78 122 C 130 122, 140 76, 176 76"]} />
|
|
<Tile x={36} y={10} size={40} src="/images/notion-icon.svg" logo />
|
|
<Tile x={36} y={56} size={40} src="/images/google-drive-icon.svg" logo />
|
|
<Tile x={36} y={102} size={40} src="/images/github-icon.svg" logo />
|
|
<At x={176} y={50} className="m-orb is-hub"><img src="/images/art/mark-white.svg" alt="" /></At>
|
|
</ArtTile>
|
|
<ArtTile title="Extractors" href="/concepts/content-types" art="/images/art/bg-islands.jpg" pos="center">
|
|
<At x={22} y={36} className="m m-card m-pdf"><Icon icon="/icons/hugeicons/file-02.svg" size={20} /><span className="m-pdf-tag">PDF</span></At>
|
|
<Lines d={["M90 76 L 136 76"]} />
|
|
<At x={142} y={30} w={100} className="m m-card m-chunk is-faded"><span className="m-skel" style={{ width: "70%" }} /></At>
|
|
<At x={142} y={62} w={100} className="m m-card m-chunk is-on"><span className="m-skel" style={{ width: "84%" }} /></At>
|
|
<At x={142} y={94} w={100} className="m m-card m-chunk is-faded"><span className="m-skel" style={{ width: "58%" }} /></At>
|
|
</ArtTile>
|
|
<ArtTile title="Qualitative analysis" href="https://supermemory.ai/research" art="/images/art/bg-meadow.jpg" pos="30% 60%" flip>
|
|
<At x={42} y={10} w={180} className="m m-card m-board">
|
|
<div className="m-board-head">LongMemEval</div>
|
|
<div className="m-board-row is-on"><span className="m-ic is-on"><Icon icon="/icons/hugeicons/award-01.svg" size={12} /></span><span className="m-grow">Supermemory</span><span>1st</span></div>
|
|
<div className="m-board-row is-faded"><span className="m-ic" /><span className="m-grow"><span className="m-skel" style={{ width: "64%" }} /></span></div>
|
|
<div className="m-board-row is-faded"><span className="m-ic" /><span className="m-grow"><span className="m-skel" style={{ width: "48%" }} /></span></div>
|
|
</At>
|
|
</ArtTile>
|
|
<ArtTile title="Plugins" href="/supermemory-mcp/mcp" art="/images/art/bg-lakeshore.jpg" pos="70% 40%" flip>
|
|
<Tile x={20} y={52} size={48} src="/images/claude-code-icon.svg" logo on badge />
|
|
<Tile x={82} y={52} size={48} src="/images/cursor-logo.svg" badge />
|
|
<Tile x={144} y={52} size={48} src="/images/codex.svg" badge />
|
|
<Tile x={206} y={52} size={48} src="/images/openclaw-logo.jpg" badge />
|
|
</ArtTile>
|
|
</div>
|
|
|
|
With Supermemory, your agent remembers what each user has told it and uses that to answer more personally and more consistently.
|
|
|
|
Supermemory leads the LongMemEval and LoCoMo benchmarks and independent ones such as SWEContext. See the [benchmark results](https://supermemory.ai/research).
|
|
|
|
## How does it work? (at a glance)
|
|
|
|
|
|
|
|
- You send raw data in any format (text, files and chats), or connect a data source.
|
|
- Supermemory indexes it and builds a graph of what it learns about each entity: a user, a document, a project or an organization. Each entity is identified by a `containerTag`.
|
|
- Your agent searches that graph for memory or retrieval, and Supermemory keeps a profile of each entity up to date.
|
|
|
|
For the full pipeline, read [how ingestion works](/concepts/how-it-works).
|
|
|
|
## 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 a lasting understanding of people and entities over time: their preferences, decisions, relationships and corrections. Retrieval (RAG) grounds answers in documents and knowledge bases. Most agents need both.
|
|
|
|
With memory, your agent can:
|
|
|
|
- Personalize answers with preferences, roles and history from earlier sessions, without putting the whole chat log in every prompt.
|
|
- Stay correct when facts change. If a user says "I love Adidas" and later "I'm switching to Puma", only the newer preference should hold.
|
|
- Pull the right policy, ticket or document when a question needs source material.
|
|
- Keep each customer's memory separate, so one user's data never leaks into another's.
|
|
|
|
Think of memory as the context a good teammate carries in their head, not a search box over raw logs. To see where retrieval ends and memory begins, read [Memory vs RAG](/concepts/memory-vs-rag).
|
|
|
|
## Why Supermemory?
|
|
|
|
- **It leads long-horizon memory benchmarks.** First on LongMemEval, LoCoMo and ConvoMem, and strong on independent benchmarks like SWEContext.
|
|
- **Memory is a graph.** Facts update, connect and expire in real time instead of sitting as nearest-neighbor chunks.
|
|
- **User profiles are built in.** Static and dynamic facts about each user are ready to drop into the prompt.
|
|
- **Memory and RAG share one engine.** Personalization and document search run on the same `containerTag`.
|
|
- **Every entry point uses one store.** The API, MCP, plugins, SMFS and connectors all read and write the same memories.
|
|
- **It is multimodal by default.** Text, chats, PDFs, images, video and code are extracted automatically, whether you upload them or sync them.
|
|
- **It runs where you need it.** Use the managed cloud, or self-host a single binary that also works offline.
|
|
|
|
|
|
|
|
<Note>
|
|
Memory, profiles and SuperRAG share one context pool when they use the same `containerTag`. A container can be any scope you choose: a user, a project, a team or an organization.
|
|
</Note>
|
|
|
|
## Next steps
|
|
|
|
<CardGroup cols={2}>
|
|
<Card title="Quickstart" icon="/icons/hugeicons/play.svg" href="/quickstart">
|
|
Make your first API call in minutes
|
|
</Card>
|
|
<Card title="How it works" icon="/icons/hugeicons/cpu.svg" href="/concepts/how-it-works">
|
|
Understand the knowledge graph architecture
|
|
</Card>
|
|
<Card title="Comparison" icon="/icons/hugeicons/balance-scale.svg" href="/overview/comparison">
|
|
vs DIY vectors, thin memory layers, pure RAG
|
|
</Card>
|
|
<Card title="Self-host it" icon="/icons/hugeicons/server-stack-01.svg" href="/self-hosting/overview">
|
|
One binary, zero config, fully offline
|
|
</Card>
|
|
<Card title="Billing & plans" icon="/icons/hugeicons/credit-card.svg" href="/overview/billing">
|
|
Credits, SM tokens, and how usage works
|
|
</Card>
|
|
<Card title="Security & compliance" icon="/icons/hugeicons/shield-01.svg" href="/overview/security">
|
|
SOC 2, GDPR, HIPAA BAA, encryption
|
|
</Card>
|
|
</CardGroup>
|