MemMachine
https://github.com/MemMachine/MemMachine
Multi-layered open-source memory system for AI agents.
Active · Verified 2026-10-22 · fresh as of 2026-10-22 · 48 cited facts, 1 with a third-party source · 1 recorded change · Suggest a correction
At a glance
- Type
- Multi-layered memory system repo fix
- Tier
- T2
- Section
- Dedicated memory layers
- Latest release
- v0.3.9 (2026-05-18) registry fix
- License
- Apache-2.0 repo fix
- GitHub
- MemMachine/MemMachine — ~3.2k stars repo fix
- Pricing
- OSS free (Apache-2.0); managed MemMachine Platform with free account (paid tiers not listed in README) repo fix
- Funding
- searched not found fix
- Freshness
- 2026-10-22 fix
- Layer
- agent-infra repo fix
- SDK languages
- python, rest, typescript repo fix
- Protocols
- mcp-server repo fix
- Developer activity
- OpenRank 6.66 (12-mo avg 16.39) · 19 participants — 2026-09 third-party fix
Memory taxonomy
- storage
- vector
- retrieval
- similarity
- persistence
- long-term
- update
- extraction
- unit
- episode
- governance
- inspectable
- conflict
- none
When to use
Optimised for: multi-layer memory architecture
Anti-fit: no anti-fit explicitly stated
Pros & cons
Pros
OSS framework for personalized memory across agents; multi-LLM support and clean SDK.
Cons
Distinct from the MemMachine paper above — naming overlap can confuse comparison.
Claims & capabilities
Apache-2.0; multi-layered memory architecture; v0.3.7 released 2026-05-01; first-party MemMachine MCP; 0.9169 LoCoMo (paper); paper arXiv:2604.04853 also titled MemMachine; Enterprise version forthcoming
Technical surface
- API surface
- REST, SDK: Python
- Backend storage
- hybrid
- Deployment
- Both — self-hosted, or managed MemMachine Platform
- Embedding model
- multiple supported
- Multi-tenancy
- namespace
- MCP
- native (first-party) — MemMachine MCP
- A2A
- not documented publicly
- OpenTelemetry
- not documented publicly
Alternatives to MemMachine → live options ranked by shared design, licence, hosting and MCP support
Similar systems
Other dedicated memory layers in the catalog, ranked by inbound references.
- Mem0 T1
Universal memory layer for AI agents. Three concurrent stores (vector + graph + KV); LLM-extracted facts; concurrent retrieval via ThreadPoolExecutor.
- Zep & Graphiti T1
Bi-temporal knowledge graph (event time + ingestion time). Strong on chronological reasoning and contradiction tracking. Graphiti is the open-source core.
- Cognee T1
"Extract–Cognify–Load" pipeline that turns raw input into a typed, queryable knowledge graph for agent recall.
- Hindsight (Vectorize) T1
Standalone memory service from Vectorize. Open source. Biomimetic four-network design (World, Bank, Observation, Opinion). Ships an MCP memory server.
- Letta / MemGPT T1
Originated as MemGPT (UC Berkeley), where the model pages its own memory between context, recall and archival tiers. The product now centres on Letta Code, a stateful agent harness whose agents rewrite their own memory, skills and prompts over time.
- Memori T2
Structured, persistent memory layer for AI agents. Open source.
Sources (47 cited facts, 1 third-party)
Each fact links to the page it came from; the badge says who published it (how to read sources).
- What it is Multi-layered memory system repo
- What it does & what's distinct Multi-layered open-source memory system for AI agents. repo
- Claims / benefits Apache-2.0; multi-layered memory architecture; v0.3.7 released 2026-05-01; first-party M… repo
- Modalities text repo
- Created 2025-08 repo
- Latest release v0.3.9 (2026-05-18) registry
- License Apache-2.0 repo
- GitHub MemMachine/MemMachine — ~3.2k stars repo
- Mindshare 3.1k GitHub stars; +30/mo growth; below detection threshold for HN/press; 2 academic cit… repo
- Citations 2 cites 2/yr paper
- Pricing OSS free (Apache-2.0); managed MemMachine Platform with free account (paid tiers not lis… repo
- Deployment Both — self-hosted, or managed MemMachine Platform repo
- Founders / pedigree MemVerge (initial incubator); no named individual founders publicly identified vendor
- Memory volume / scale no scale claim repo
- Performance 0.9169 LoCoMo F1 (paper Table 4; SOTA at submission) paper
- Reproducibility Too recent paper
- Code/weights release Code public (Apache-2.0) repo
- API surface REST, SDK: Python repo
- Backend storage hybrid repo
- Multi-tenancy namespace repo
- Embedding model multiple supported repo
- Schema evolution schema with migration repo
- Namespace primitives per-tenant repo
- Contradiction handling LLM resolves and overwrites repo
- MCP support native (first-party) — MemMachine MCP repo
- Integration count 0 repo
- Orchestration both (per-agent personas; shared episodic graph) vendor
- Programmatic control full CRUD via lib vendor
- Pricing specifics OSS free; Enterprise coming soon — no public price vendor
- Validated verticals agent developers (horizontal) vendor
- Time to running <1hr (OSS Docker) vendor
- Anti-fit no anti-fit explicitly stated vendor
- Optimised for multi-layer memory architecture vendor
- Adjacent infrastructure BYO LLM; bundled stores vendor
- Pros OSS framework for personalized memory across agents; multi-LLM support and clean SDK. repo
- Cons Distinct from the MemMachine paper above — naming overlap can confuse comparison. repo
- Project & sources GitHub MemMachine/MemMachine repo
- commit-trajectory [{"ym":"2025-08","cum":24},{"ym":"2025-09","cum":344},{"ym":"2025-10","cum":476},{"ym":"… repo
- cost-pricing-model subscription vendor
- cost-last-verified 2026-06-29 vendor
- architecture-deployment-locus hybrid repo
- licensing-bucket open-core vendor
- positioning-platform-vs-product platform repo
- taxonomy-three-layer-rollup agent-infra repo
- developer-activity-opendigger {"activity":80.26,"as_of":"2026-09","openrank_12mo":16.39,"openrank_current":6.66,"parti… third-party
- sdk-languages python,rest,typescript repo
- protocol-support mcp-server repo
What changed
Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)
- 2026-10-22
- Latest release “v0.3.7 2026-05-01” → “v0.3.9 (2026-05-18)” source
- Pricing “OSS free; Enterprise version coming soon (no public price)” → “OSS free (Apache-2.0); managed MemMachine Platform with free account (paid tiers not listed in README)” source
- Deployment “Self-hosted (private cloud or on-premises); Enterprise managed forthcoming” → “Both — self-hosted, or managed MemMachine Platform” source