Supermemory
Memory and context infrastructure for AI agents, combining managed retrieval, broad content ingestion, and connectors that vary by plan.
About Supermemory
Supermemory gives AI agents a shared memory and context layer. Developers send text, conversations, files, images, documents, or videos; Supermemory extracts and indexes the material, builds semantic understanding around a user, project, document, or organization, then retrieves relevant context when the agent needs it. Agent memory, content extraction, connectors, syncing, and managed RAG sit in one service that works with any model. For a development team, that means less retrieval infrastructure to assemble and more continuity between sessions. The tradeoffs are practical: usage beyond each plan's allowance adds cost, connectors vary by tier, and teams still need to assess how uploaded and synchronized data fits their governance rules. The Free plan includes about $5 of monthly usage. Paid tiers add larger allowances, more connectors, team functions, support, and production controls.
Best for
Best use cases
Add persistent user and project memory to an AI agent.
Index mixed content and retrieve relevant context for model calls.
Sync knowledge from supported productivity or storage services.
Compare before choosing
Compare included monthly usage against expected query and ingestion volume.
Check whether required connectors are available on the intended plan.
Evaluate support, compliance, spend controls, and deployment options before production use.
Features
Agent memory
Store short-term and long-term memory so an AI agent can recall user and project context across interactions.
Content ingestion
Send text, conversations, files, images, documents, and videos for extraction and semantic indexing.
Relevant context retrieval
Query the indexed memory and pass only the most relevant context to the model at response time.
Connectors and syncing
Sync external knowledge through connectors, with availability depending on the selected pricing plan.
Managed RAG
Use a managed retrieval layer and semantic understanding graph instead of assembling the full retrieval stack yourself.
Pros
- The Free plan includes about $5 of monthly usage, which lowers the cost of prototyping.
- Support for text, conversations, files, images, documents, and videos reduces the need for separate ingestion pipelines.
- The service works with any model, so teams are not tied to one model provider.
- Paid plans provide unlimited storage and users, with higher tiers adding production controls and compliance options.
Cons
- Usage beyond the amount included with each plan can add variable cost, so production teams need to monitor spend.
- Connector access depends on the plan, which may require an upgrade for Gmail, GitHub, S3, or other sources.
- The Free plan relies on community support, while direct and priority support require paid tiers.
- The Scale plan costs $399 per month, a substantial starting commitment for teams that need compliance or self-hosting options.
Pricing Plans
Free
- About $5 per month of usage included
- Hermes Plugin
- Supermemory MCP
- Community support
Pro
- About $20 per month of usage included
- Unlimited storage
- Unlimited users
- Google Drive, Notion, and OneDrive connectors
- 2 teammates included
- OpenClaw, Claude Code, and other plugins
- Auto top-up available
- Email support
Max
- About $130 per month of usage included
- Unlimited storage
- Unlimited users
- Gmail and Granola connectors, plus Pro connectors
- OpenClaw, Claude Code, and other plugins
- Auto top-up available
- Priority support
Scale
- About $600 per month of usage included
- Unlimited storage
- Unlimited users
- Up to 10 teammates
- All connectors, including Gmail, GitHub, S3, Web Crawler, and Pro connectors
- Auto top-up and spend caps
- Priority support
- SOC 2 and HIPAA BAA
- Self-hosted option
* Pricing information is for reference only. Please refer to the latest announcement on Supermemory's official website.