Persistent memory API for AI agents
Persistent memory for AI agents
One memory. Every AI tool. Write once, recall anywhere — across Claude Code, Cursor, VS Code, ChatGPT, and Python, with a single API key.
claude mcp add mnemoverse \
-e MNEMOVERSE_API_KEY=mk_live_YOUR_KEY \
-- npx -y @mnemoverse/mcp-memory-server@latestor let Claude set it up for you
claude "Add the @mnemoverse/mcp-memory-server MCP server with env var MNEMOVERSE_API_KEY, then ask me for my key"Mnemoverse is a persistent memory API for AI agents — write a preference, decision, or lesson once and recall it from any tool. It is not a vector database: memories learn from outcomes via Hebbian associations and Rescorla-Wagner, consolidate and forget via HDBSCAN and Von Restorff, and improve with use. One API key connects six client integrations (Claude Code, Claude Desktop, Cursor, VS Code, Windsurf, ChatGPT) plus Python and REST, and an MCP server exposing six tools. The free tier is $0 with 1,000 queries/day. Mnemoverse is #2 on the public LoCoMo leaderboard, and the underlying method is described in the SLoD paper (arXiv:2603.08965), presented at the GRAAI workshop, IEEE WCCI 2026.
Not a vector database
Memory that learns, forgets, and improves
Vector databases store and retrieve. Mnemoverse learns which memories matter — Hebbian associations link related concepts, outcome feedback tunes ranking, and consolidation keeps memory dense instead of unbounded.
| Mnemoverse | Vector DB (Pinecone / Weaviate / Chroma) | |
|---|---|---|
| Core model | Statistical learning (Rescorla-Wagner + Hebbian) | Vector embeddings (cosine similarity) |
| Learns from outcomes | Yes — feedback updates valence | No outcome-feedback loop (reranking is similarity-based) |
| Concept associations | Three-factor Hebbian graph | None |
| Memory compression | HDBSCAN consolidation + Von Restorff | Accumulate forever |
| Query expansion | Automatic via learned associations | No learned-association expansion (hybrid/keyword only) |
| Starting price | Free (1,000 queries/day) | Free tier, then ~$20–50/mo |
How it works
Write → Learn → Recall
The same loop on every tool. Memories that lead to good outcomes rank higher next time. The system improves with use.
Write
Store a preference, decision, or lesson with a single call. An importance gate filters noise before anything is kept, so memory stays signal, not scratch.
Learn
Report outcomes and the system updates valence. Memories that lead to good answers rank higher; ones that consistently fail get suppressed. Consolidation merges look-alikes.
Recall
Query in natural language. Hebbian expansion pulls in associated concepts and valence-boosted ranking surfaces what actually mattered last time — across every tool, one API key.
See it work
Watch an agent's memory take shape
This is a real memory graph, not an illustration — atoms, associations, and consolidation as the engine stores and recalls. Drag to explore.
Atoms are memories
Each node is one stored memory — sized by importance, colored by type. The graph is what the agent actually knows.
Links are associations
When two concepts are recalled together, a Hebbian link forms and thickens — so related memories surface as a group, not one at a time.
Consolidation keeps it dense
Look-alike memories merge over time (HDBSCAN), while distinctive ones are protected — recall stays sharp instead of drowning in noise.
Retrieval lights the path
A query activates associated concepts and walks the links to the right atom — the same recall that put Mnemoverse #2 on the LoCoMo leaderboard.
For agents (MCP)
One line to wire memory into your agent
The @mnemoverse/mcp-memory-server npm package exposes six tools — memory_write, memory_read, memory_feedback, memory_stats, memory_delete, memory_delete_domain — to any MCP-compatible client: Claude Code, Claude Desktop, Cursor, VS Code + Copilot, and Windsurf. Listed on the official MCP Registry.
claude mcp add mnemoverse \
-e MNEMOVERSE_API_KEY=mk_live_YOUR_KEY \
-- npx -y @mnemoverse/mcp-memory-server@latestBenchmarks
Benchmarked, honestly
Mnemoverse is evaluated on public memory benchmarks — LoCoMo and LongMemEval — rather than internal scores you can't reproduce. On the public LoCoMo leaderboard, Mnemoverse currently sits at #2.
Live, verified numbers: see the benchmarks page.
Pricing
Start free. Scale when you need to.
Free tier ships with full API access and the MCP server — no credit card required.
Enterprise
Unlimited atoms & queries · SSO/SAML · audit logs · SLA · data residency · BAA/SOC 2 on request.
Contact salesFAQ
Frequently asked
It learns from outcomes (Hebbian associations + Rescorla-Wagner), consolidates and forgets stale memory (HDBSCAN + Von Restorff), and improves over time. A vector database only does similarity search.
Yes. One API key works across Claude Code, Claude Desktop, Cursor, VS Code, Windsurf, ChatGPT, Python, and any HTTP client.
Yes. Free is $0: 1,000 queries/day, 10,000 atoms, 60 req/min — no credit card.
Per-tenant org isolation. API keys are SHA-256 hashed and compared in constant time, and we don't train on your data.
RAG answers “what do the docs say?” Mnemoverse remembers “what did we discuss, decide, or learn last time?” They're complementary.
Yes. @mnemoverse/mcp-memory-server exposes six tools and installs with a one-line `claude mcp add`.
Get started
Start building
Sign up, grab a free API key, and give your agents memory in minutes.