Supermemory: Give agents memory that keeps learning
Supermemory is a memory and continual learning layer for AI agents, delivered through an API, plugins and MCP.
What is Supermemory?
Supermemory is a memory layer for AI agents. Raw data such as files, chats, URLs and other input enters the system, where the company's learner-1 model updates, merges, infers and forgets against a memory database, and the result is injected back into an agent's context in real time. The company says learning and memory sit outside model providers so they carry across models and harnesses. The product is offered through an API, plugins and MCP, and can be deployed in a data center, a VPC, on a laptop or in air-gapped environments. Billing runs on SM tokens, which measure unique content ingested; repeats and unchanged content are not billed again on the token meter. The company states it is SOC 2 certified, HIPAA compliant and GDPR aligned.
Focus areas: Agent Memory, MCP.
Who is Supermemory for?
- Developers building AI agents that need persistent context
- Teams running production agent workloads at scale
- Organizations with data residency or air-gapped deployment needs
How much does Supermemory cost?
Supermemory starts at $19/mo. Always check the vendor’s own pricing page before you buy.
| Plan | Price | What you get |
|---|---|---|
| Free | $0 per month | $5 of credits included, renewed every month |
| Pro | $19 per month | $20 included per month; 3 team seats, unlimited end users |
| Max | $100 per month | $130 included per month; 6.5× the credits of Pro; Gmail connector |
| Scale | $399 per month | $600 included per month; unlimited team seats, per-tag access |
Is Supermemory worth it?
Supermemory targets a real gap in agent infrastructure, but buyers should weigh its self-reported benchmarks and plan-gated features against their own workloads
Where Supermemory is strong
- Memory and learning are kept separate from model providers, so they carry across models and harnesses
- Deployment options include a data center, a VPC, a laptop and air-gapped setups
- Billing only charges for unique content, with repeats and unchanged content free on the token meter
Where it falls short
- No independent verification of the published benchmark and latency figures is provided in the source
- Advanced PDF extraction is limited to the Scale and Enterprise plans
How do you get started with Supermemory?
- 01Create an account and start on the free plan, which includes $5 of monthly credits
- 02Copy the setup prompt or connect through the API, plugins or MCP
- 03Send data into the memory store and query it from your agent's context
Supermemory: frequently asked questions
How does billing work?
Every plan includes a monthly dollar balance of credits. Each API call, whether storing, searching or indexing, draws from that balance at the published rates, and only unique content is billed.
Can I self-host Supermemory?
Self-hosted deployments are available on the Scale and Enterprise plans. Enterprise also supports fully air-gapped deployments, where LLM inference may be the only outbound call or none at all.
Do unused credits roll over?
Subscription credits reset monthly. Credits bought as a top-up never expire and stay in your balance until used.
Is there a free option for startups or researchers?
Qualifying early-stage startups and academic research teams can get the Scale plan free for three months, including its included usage and dedicated support.
Supermemory alternatives
The closest tools in our directory, matched on category, focus areas and audience.
| Tool | Best for | Free plan | Starting price | Focus areas |
|---|---|---|---|---|
| Supermemory | Developers adding persistent, cross-model memory to AI agents | Yes | $19/mo | Agent Memory, MCP |
| Versuno | Unify all your AI assets in one place. | Yes | — | Agent Memory |
| Search Candy | Developers needing citable, versioned document passages for AI agents | No | — | Agent Memory |
| Inflectiv.ai | Inflectiv.ai: AI-driven platform for data ownership, validation, and monetization in a decentralized ecosystem. | Unknown | — | MCP |
| Vectorize | Teams giving LLM agents memory that improves with use | Yes | — | Agent Memory, Vector Search |
| Cognee | Cognee is an AI platform that builds a dynamic, self-organizing knowledge graph from enterprise data, enabling deep contextual… | Unknown | — | Agent Memory |
| Zep | Chatbot platform with language comprehension and memory. | Unknown | — | Agent Memory |
| Ducky | Developers adding managed search and RAG to AI products | Yes | — | — |
| Memable | Personal knowledge organization & retrieval. | Yes | — | — |
| Julep AI | Developers automating multi-step AI and browser tasks | Unknown | — | — |
| xmem | Xmem centralizes company knowledge, streamlines sharing, and integrates with APIs & LLMs. | Unknown | — | — |

