# Supermemory

> The memory layer for AI agents. Context engineering platform powering enterprise APIs, developer plugins, and a personal app that remembers everything.

Supermemory is a research and product lab building the best context and memory infrastructure for AI agents. We provide five layers of context: User Profiles, Memory Graph, Retrieval, Extractors, and Connectors — all in one API.

## Quick Start

- [API Documentation](https://docs.supermemory.ai)
- [TypeScript SDK](https://supermemory.ai/docs/integrations/supermemory-sdk)
- [Python SDK](https://supermemory.ai/docs/integrations/supermemory-sdk)
- [REST API Reference](https://supermemory.ai/docs/api-reference/manage-documents/add-document)
- [Developer Console (API Keys)](https://console.supermemory.ai)
- [Personal App](https://app.supermemory.ai)
- [SuperRAG as a Filesystem](https://smfs.ai)

## Core API Endpoints

- Add memories: POST https://api.supermemory.ai/v3/add
- Search memories: POST https://api.supermemory.ai/v3/search
- Full reference: https://docs.supermemory.ai/api-reference

## Integrations

Works with: TypeScript, Python, REST API, Claude Code, Cursor, OpenClaw, OpenCode, Windsurf, Vercel AI SDK, LangChain, LangGraph, CrewAI, OpenAI SDK, Mastra, Zapier, n8n, Pipecat

- [All integrations](https://supermemory.ai/docs/integrations/supermemory-sdk)

## Key Features

### User Profiles
Builds deep user profiles from behavior — intent, preferences, and context. Automatically extracts and maintains structured facts about users from their interactions. Profiles update continuously as new information arrives.

### Memory Graph
Custom vector graph engine with ontology-aware edges. Knowledge updates, merges, contradicts, and infers. Unlike simple vector stores, the memory graph understands relationships between memories and can resolve contradictions when information changes over time.

### Retrieval
Hybrid vector + keyword search with sub-300ms latency p50. Context-aware reranking ensures the most relevant memories surface first. Supports filtering by user, time range, and custom metadata. Also accessible via the SuperRAG filesystem at https://smfs.ai.

### Extractors
Understands any format — PDFs, web pages, images, audio. Smart chunking preserves meaning across document boundaries. Automatically extracts structured data from unstructured content. SOTA extraction quality is continuously tracked.

### Connectors
Notion, Google Drive, S3, Gmail, and custom sources. Auto-sync keeps memories up to date as source data changes. Configure once, memories stay current.

## Benchmarks

- LongMemEval: 85.2%
- LoCoMo: #1
- ConvoMem: #1
- MemoryBench (our open eval platform): https://git.new/membench
- Research paper: https://supermemory.ai/research

## Pricing

Same plans whether you sign up via the Developer Console (API/SDK) or the Personal App — toggle on the pricing page picks the right destination. Every plan includes monthly usage credits; pay only for what you use beyond that. SM tokens are billed only on unique content (100% prompt-cache discount, baked in).

### Free — $0/mo

For builders tinkering, prototypes and side projects.

- ~$5/mo of usage built in
- Hermes Plugin
- Supermemory MCP
- Community support

[Start free (Developer)](https://console.supermemory.ai/billing) · [Start free (Personal app)](https://app.supermemory.ai/settings#account)

### Pro — $19/mo

For small teams and plugin power-users.

- ~$20/mo of usage built in
- Unlimited storage
- Unlimited users
- Auto top-up available
- Google Drive, Notion & OneDrive connectors
- 2 teammates included
- OpenClaw, Claude Code and other plugins
- Email support

[Get Pro (Developer)](https://console.supermemory.ai/billing) · [Get Pro (Personal app)](https://app.supermemory.ai/settings#account)

### Max — $100/mo

More headroom for developers who need it.

- ~$130/mo of usage built in (6× Pro)
- Unlimited storage
- Unlimited users
- Gmail & Granola connectors (+ Pro)
- Auto top-up available
- OpenClaw, Claude Code and other plugins
- Priority support

[Get Max (Developer)](https://console.supermemory.ai/billing) · [Get Max (Personal app)](https://app.supermemory.ai/settings#account)

### Scale — $399/mo

For teams running production workloads.

- ~$600/mo of usage built in
- Unlimited storage
- Unlimited users
- Up to 10 teammates
- All connectors (Gmail, GitHub, S3, Web Crawler + Pro)
- Auto top-up + spend caps
- Priority support
- SOC 2 · HIPAA BAA
- Self-hosted option

[Get Scale (Developer)](https://console.supermemory.ai/billing) · [Get Scale (Personal app)](https://app.supermemory.ai/settings#account)

### Enterprise — Custom

For organizations with committed spend, custom deployments, and security requirements.

**Security & compliance**

- Air-gapped self-hosting, you own the data.
- Or a dedicated managed instance
- SOC 2 · HIPAA · GDPR
- Custom contracts & DPA
- Enterprise MCP

**Scale & performance**

- ~Unlimited usage
- Committed-spend pricing
- Custom rate limits & throughput
- Dedicated infrastructure

**Enterprise-grade support**

- Dedicated account manager
- Forward Deployed engineer
- 1:1 onboarding & integration
- Uptime SLA
- Priority Slack channel

[Talk to us](https://cal.com/supermemory/30min?duration=45&overlayCalendar=true)

## Rate Card

Same per-unit rates across Free, Pro, and Scale. Enterprise unlocks committed-spend discounts. Billed in SM tokens — only unique content we ingest is billed. Repeats cost nothing.

### Memory

Memory graph per user. Auto profiles and fact hierarchies so agents learn in real-time.

- Plain text: $0.005 /1K SM tokens
- Rich content: $0.010 /1K SM tokens

2× cheaper than next-best, with better quality. Powered by our own model.

### SuperRAG

Multimodal Extraction -> Contextual Chunking -> Retrieval for agents. No embeddings or vectors required.

- Text mode: $0.001 /1K SM tokens
- Rich mode: $0.002 /1K SM tokens

[Available as a filesystem](https://smfs.ai)

'Rich' = images, PDFs, audio, video. SOTA extraction, always tracked.

### Search and Traversal

Semantic search and graph traversal across your stored content, in a single call.

- Per query: $0.005 /1K queries

Features:
- Hybrid search: RAG + Memory in one call
- Graph traversal across linked memories
- Configurable filters and re-ranking

Sub-300ms p50. Built for agent loops

### Operations

Additional operations for API calls

- Per operation: $0.10 /1K operations

Features:
- Re-ranking
- Aggregation
- Query rewriting
- Other operations

Composable building blocks for richer queries.

## Startup & Research Program

Qualifying early-stage startups and academic research projects get production-grade memory infrastructure, free. Ship the agent, we'll cover the context.

- up to $2,000 in free credits
- Dedicated support
- 3 months to build
- All features unlocked

[Apply now](https://binary.so/eQvzzEk)

## Pricing FAQ

### How does the credit balance work?

Every plan comes with a monthly dollar balance. Each API call — storing memories, searching, indexing — draws from that balance at the rates listed above. When the balance runs out, you can top up or auto top-up to keep going. No surprise bills at the end of the month.

### What's an SM token?

SM tokens are the unique tokens Supermemory actually ingests and embeds — repeats and unchanged content don't get billed again. Effectively a 100% discount on what a normal prompt cache would re-charge you for. Plain text is $0.005 per 1K SM tokens; rich content (PDFs, audio, video) is $0.010 per 1K because it needs heavier extraction.

### What happens if I re-store the same content?

Nothing — you're not billed again. Supermemory deduplicates at the token level, so re-uploading a doc, syncing a connector, or pushing the same conversation history won't redraw from your balance. Only net-new content counts as SM tokens. This is why production agents that loop over the same context end up an order of magnitude cheaper here than with a typical vector DB.

### Do unused credits roll over?

Subscription credits reset monthly. Top-up credits you purchase in advance never expire — they sit in your balance until you use them.

### What happens if I exceed my balance?

On Free, you'll be paused: pay-as-you-go isn't available on Free, so you'll need to upgrade to a paid plan to keep going. On Pro, Max and Scale, auto top-up kicks in to keep your app running. You can set hard spend caps on Scale to prevent runaway usage.

### Can I self-host?

Self-hosted deployments are available on Scale and Enterprise. Enterprise additionally supports fully air-gapped deployments (LLM inference may be the only outbound dependency - Unless GPUs are available!).

### Do you offer startup or academic research credits?

Yes — qualifying early-stage startups and academic research teams get up to $2,000 in credits, dedicated support, and 3 months to build. Apply via the Startup Program link above.

## Personal Supermemory

A consumer app at https://app.supermemory.ai — one memory across all your AI tools. Plugins for Claude Code, Cursor, OpenClaw, OpenCode, Windsurf, and a Chrome extension. 10,000+ power users. Same plans as the developer console; billing managed at https://app.supermemory.ai/settings#account.

## Company

- Website: https://supermemory.ai
- GitHub: https://github.com/supermemoryai
- X/Twitter: https://x.com/supermemory
- Security: https://supermemory.link/security
- Contact: https://cal.com/supermemory/30min?duration=45&overlayCalendar=true

## Pages

- [Home](https://supermemory.ai/)
- [Connectors](https://supermemory.ai/connectors): Data source integrations — Notion, Google Drive, Gmail, S3, and more
- [MCP](https://supermemory.ai/mcp): Model Context Protocol — universal memory for every AI tool
- [Plugins](https://supermemory.ai/plugins): IDE and AI tool plugins — Claude Code, Cursor, OpenClaw, OpenCode, Windsurf
- [RAG](https://supermemory.ai/rag): Retrieval-Augmented Generation — hybrid search with sub-300ms latency
- [Memory Graph](https://supermemory.ai/memory-graph): Living knowledge graph — automatic fact extraction, intelligent relationships, and built-in forgetting
- [Pricing](https://supermemory.ai/pricing): Plans, features, and pricing details
- [Pricing (Markdown)](https://supermemory.ai/pricing.md): Machine-readable pricing
- [Research](https://supermemory.ai/research): Benchmarks and research papers
- [Consumer App](https://supermemory.ai/consumer): Personal memory app
- [SuperRAG Filesystem](https://smfs.ai): SuperRAG accessed as a mounted filesystem
- [Privacy Policy](https://supermemory.ai/privacy-policy)
- [Terms of Service](https://supermemory.ai/terms-of-service)
- [API Docs](https://docs.supermemory.ai)
- [Developer Console](https://console.supermemory.ai)
- [Personal App](https://app.supermemory.ai)