[Introducing supermemory Read the post ](https://x.com/supermemory/status/2097035274094272935) 

# supermemory is building the default engine for memory and continual learning for agents.

Available through our API, plugins, and MCP.

[Build with supermemory ](https://console.supermemory.ai/) [Talk to sales](https://cal.com/supermemory/meet?overlayCalendar=true)

Works with

* [![ChatGPT](https://supermemory.ai/logos/tools/chatgpt.svg) ](https://supermemory.ai/docs/supermemory-mcp/chatgpt-web "ChatGPT")
* [![Claude](https://supermemory.ai/logos/tools/claude-desktop.svg) ](https://supermemory.ai/docs/supermemory-mcp/claude-desktop "Claude")
* [![Cursor](https://supermemory.ai/logos/tools/cursor.svg) ](https://supermemory.ai/docs/integrations/cursor "Cursor")
* [![Hermes](https://supermemory.ai/logos/tools/hermes.svg) ](https://supermemory.ai/docs/integrations/hermes "Hermes")
* [![MCP](https://supermemory.ai/logos/tools/mcp.svg) ](https://supermemory.ai/docs/supermemory-mcp/mcp "MCP")

The path to useful AI is to scale [in-context continuous learning](https://supermemory.ai/#q-hassabis) that works with **any model, any harness**, and for all the varieties of use cases. As open-source model adoption grows and the world becomes even more [multi-model](https://openrouter.ai/rankings), learning and memory should be separated from the model providers and be highly interoperable (hence in-context!).

[Intelligence is no longer a constraint.](https://arcprize.org/leaderboard)

Pre-training and post-training, scaling on data and compute, will continue to push models’ intelligence, but will not enable them to learn and improve in real time.

We build all the hard infrastructure parts of a scalable memory system, with [dense, interconnected, growing learnings](https://supermemory.ai/docs/concepts/graph-memory), and an understanding of time.

Our model **learner-1** helps any model continuously improve by extracting and dreaming on the context of any user, task, or tenant, and putting it on our vector-graph database. This is then brought to the model by injecting tokens into its context in real time.

fileschatsurlsraw datalearner-1memory dbv12prior memories in contextprior memoriesinject via hooksyour agent's contextor[![Claude Code](https://supermemory.ai/logos/tools/claude-code.svg) ](https://supermemory.ai/docs/integrations/claude-code) [![Cursor](https://supermemory.ai/logos/tools/cursor.svg) ](https://supermemory.ai/docs/integrations/cursor) [![MCP](https://supermemory.ai/logos/tools/mcp.svg) ](https://supermemory.ai/docs/supermemory-mcp/mcp)

Figure 1\. Raw data enters supermemory. learner-1 updates, merges, infers and forgets against the memory db, and the result is injected into your agent’s context.

## What we do

Memory that keeps learning.

learner-1 extracts and dreams on the context of every user, task, and tenant, stores it in a vector-graph database, and brings it back to the model by injecting tokens into its context in real time.

Any model, any harness.

Learning and memory live outside the model providers, so they carry across models and harnesses and stay interoperable, in context, and yours.

The hard infrastructure, done.

Dense, interconnected, growing learnings with an understanding of time, already serving 100k+ organizations and over a trillion tokens a month.

Scale and Enterprise include advanced document extraction for PDFs, with OCR and descriptions of figures and diagrams. Free, Pro and Max keep standard PDF extraction. [Explore document support](https://supermemory.ai/docs/concepts/content-types#advanced-document-extraction).

[Read more about the specifics](https://supermemory.ai/product)

## In production

Our **promise** is the best memory for agents: the kind people **feel**, not just the kind that tops a benchmark.

Recall latencyper query, on production traffic

|         | Server  | End to end |
| ------- | ------- | ---------- |
| Search  | 187_ms_ | 356_ms_    |
| Profile | 166_ms_ | 248_ms_    |

Tokens processedper month1T+

Organizationstens of millions of end users100k+

Independent benchmarksSWE-ContextBench, February 2026

> “Supermemory **performs best overall**, achieving FAIL\_TO\_PASS test rate of **55.95%** and the highest resolution rate of **30.30%**.”

[Read the paper](https://arxiv.org/abs/2602.08316)

Public benchmarksState of the art on LongMemEval, LoCoMo and ConvoMem#1[See the research](https://supermemory.ai/research/)

Our benchmarks

Cost per question

$7.7964% cheaper

Questions answered

81%up from 69%

Figure 1 · xAFS cost curve

#### Per-question token cost vs. corpus size

Markdown memorysupermemory

Per-question token cost across 5 to 9,988 files (log scale). Markdown memory grows \~7× over the range; supermemory grows \~4×. The 299-file corpus is a notable outlier where Markdown memory thrashes. n = 220 (fs) + 220 (smfs) trials · [smfs.ai/research/runs](https://smfs.ai/research/runs)

[The research runs](https://smfs.ai/research/runs)

> “Reduced avg response time from **40s → 12s**. Using about **40–50% fewer tokens**.”

[Founder, Chatarmin · on X ](https://x.com/saasjesus/status/2024410743135694991)

## supermemory runs everywhere, even in air-gapped environments.

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

[Talk to us for your deployment requirements](https://cal.com/supermemory/meet?overlayCalendar=true) 
1. ### In your data center  
001
2. ### In your VPC  
002
3. ### On your laptop  
003

### Enterprise security.

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.

* ![AICPA SOC 2 certified](https://supermemory.ai/logos/compliance/soc2.svg)
* ![GDPR aligned](https://supermemory.ai/logos/compliance/gdpr.svg)
* ![HIPAA compliant](https://supermemory.ai/logos/compliance/hipaa.svg)

[Trust center](https://trust.supermemory.com/) [Self-hosting in the docs](https://supermemory.ai/docs/self-hosting/overview)

## Running memory in production?

If your agents already depend on it, they should be running on the engine the benchmarks and the numbers above describe. Any model, any harness, through our API, plugins and MCP.

[Build with supermemory ](https://console.supermemory.ai/)

## Writing

1. [01 We're open sourcing the company brain. Here's how we designed the multiplayer harness ](https://supermemory.ai/blog/open-sourcing-company-brain)
2. [02 Jev changes a lot in memory & context engineering. Here's exactly how. ](https://supermemory.ai/blog/jev-memory-context-engineering)
3. [03 I reverse-engineered Instinct's memory. Here's exactly how it works ](https://supermemory.ai/blog/reverse-engineering-instinct-memory)
[Read the blog](https://supermemory.ai/blog)

## Careers

We’re a small team based in San Francisco, building the memory layer for every model and every harness. If that is the problem you want to spend the next few years on, we would like to hear from you.

[See open roles](https://binary.so/UeNYHHE)