Semantic Kernel Memory
https://learn.microsoft.com/en-us/semantic-kernel/frameworks/agent/agent-memory
Microsoft Semantic Kernel's memory plugin (Microsoft.SemanticKernel.Plugins.Memory): a plug-in approach to long-term memory in the SK agent framework (agent memory marked experimental). Semantic Kernel's README names Microsoft Agent Framework 1.0 as its enterprise-ready successor, with a migration guide.
Active · Verified 2026-10-29 · fresh as of 2026-10-29 · 62 cited facts, 2 with a third-party source · 2 recorded changes · Suggest a correction
At a glance
- Type
- Memory plugin for SK agents vendor fix
- Tier
- T2
- Section
- Framework-embedded memory
- Latest release
- python-1.45.0 (2026-10-06) registry fix
- Pricing
- OSS (MIT); no direct cost for library; pay only for Azure/model consumption when using managed Azure infrastructure vendor fix
- Freshness
- 2026-10-29 fix
- Layer
- agent-infra repo fix
- Licensing
- open-source repo fix
- SDK languages
- csharp, java, python repo fix
- Protocols
- mcp-client repo fix
- Developer activity
- OpenRank 6.89 (12-mo avg 10.92) · 75 participants — 2026-09 third-party fix
Memory taxonomy
- storage
- vector
- retrieval
- similarity
- persistence
- cross-session
- update
- overwrite
- unit
- document
- governance
- inspectable
- conflict
- overwrite
When to use
Optimised for: Microsoft enterprise integration (Azure-first)
Anti-fit: not for non-Microsoft / non-SK stacks
Pros & cons
Pros
Microsoft-backed; first-class .NET support; integrates cleanly with Azure / M365 stacks.
Cons
.NET/Azure-shaped; smaller mind-share among Python-first AI engineers.
Claims & capabilities
Memory packages (.NET, Java, Python) reached General Availability late-2024/early-2025; H1 2025 roadmap added more sophisticated context-aware abstraction; one of the most popular AI orchestration frameworks
Technical surface
- API surface
- SDK: C#, Python, Java
- Backend storage
- pluggable
- Deployment
- Both (on-prem, containers, any cloud, or Azure AI Foundry managed)
- Embedding model
- multiple supported
- Multi-tenancy
- namespace
- MCP
- via official adapter — SK MCP plugin
- A2A
- supported — SK A2A integration
- OpenTelemetry
- first-class — built-in OTel
Alternatives to Semantic Kernel Memory → live options ranked by shared design, licence, hosting and MCP support
Similar systems
Other framework-embedded memory in the catalog, ranked by inbound references.
- Spring AI ChatMemory T3
Wraps conversation memory behind an Advisor abstraction. MessageWindowChatMemory default 20-msg sliding window (preserves system messages); pluggable ChatMemoryRepository backends (in-memory, JDBC, Cassandra, Neo4j). Memory injected as cross-cutting concern at the ChatClient level.
- Flowise Memory T1
Discontinued: Flowise (acquired by Workday, Aug 2025) was sunset in 2026 — code freeze 2026-07-29, GitHub repo archived 2026-08-13, end of life 2026-08-31. Was a visual LangChain canvas with LangChain-native memory nodes (Buffer, Buffer Window, Conversation Summary Buffer) backed by an internal chat_message DB table; Buffer Window kept the last K interactions, Summary Buffer compressed older turns while keeping recent ones verbatim.
- LangGraph Persistence T2
Distinct from LangMem. Built-in checkpointer saves graph state per super-step (short-term, thread-scoped). Store System adds long-term hierarchical key-value memory across threads with optional vector search + TTL. Postgres / Mongo / Redis stores for production.
- AutoGen Memory T2
AutoGen's memory layer: ListMemory chronological context plus teachable agents that vectorise corrections; integrates with Mem0/Zep rather than building deep memory natively. Microsoft AutoGen is in maintenance mode (community managed, bug and security fixes only), and Microsoft points new projects to Microsoft Agent Framework.
- CrewAI Memory T2
Memory subsystem inside the CrewAI orchestration framework; integrates with Mem0 for the long-term tier.
- Hermes Agent memory (Nous Research) T3
Nous Research's open-source self-improving agent keeps an agent-curated MEMORY.md/USER.md notebook injected at session start, full-text (FTS5) search over past sessions that returns raw messages without LLM summarisation, and optional Honcho user modelling. one of the most-starred agent projects of 2026.
Sources (61 cited facts, 2 third-party)
Each fact links to the page it came from; the badge says who published it (how to read sources).
- What it is Memory plugin for SK agents vendor
- What it does & what's distinct Microsoft Semantic Kernel's memory plugin (Microsoft.SemanticKernel.Plugins.Memory): a p… repo
- Claims / benefits Memory packages (.NET, Java, Python) reached General Availability late-2024/early-2025; … vendor
- Modalities text vendor
- Created 2023-02 vendor
- Latest release python-1.45.0 (2026-10-06) registry
- License MIT repo
- GitHub microsoft/semantic-kernel — ~28.4k stars, last commit 2026-10 repo
- Mindshare 2701100/mo PyPI (semantic-kern 2 press ~2500 (bundled jobs third-party
- Funding Microsoft-owned open-source project; no independent funding — part of Microsoft repo
- Customers / scale Microsoft enterprise ecosystem; BMW, KPMG, Fujitsu among Azure AI Foundry users (broader… vendor
- Pricing OSS (MIT); no direct cost for library; pay only for Azure/model consumption when using m… vendor
- Compliance Inherits Azure compliance posture: SOC 2 Type II, ISO 27001, HIPAA, PCI DSS, GDPR vendor
- Data handling Runs in-process within deployer's application; Azure-hosted path: encrypted agent-to-age… vendor
- Deployment Both (on-prem, containers, any cloud, or Azure AI Foundry managed) vendor
- HQ US (Redmond, WA — Microsoft HQ) repo
- Founders / pedigree Microsoft; open-source SDK from Microsoft vendor
- Memory volume / scale no scale claim vendor
- API surface SDK: C#, Python, Java vendor
- Backend storage pluggable vendor
- Multi-tenancy namespace vendor
- Embedding model multiple supported vendor
- Schema evolution schema with migration vendor
- Namespace primitives per-tenant vendor
- Contradiction handling always overwrite (last-write-wins) vendor
- MCP support via official adapter — SK MCP plugin vendor
- A2A support supported — SK A2A integration vendor
- OTel first-class — built-in OTel vendor
- Import / export JSON; multiple connectors (Postgres, Qdrant, etc.) vendor
- Integration count 0 repo
- Orchestration single-agent (per Kernel; multi-agent via Agent Framework) vendor
- Programmatic control full CRUD via plugin API (.NET/Python/Java) vendor
- Pricing specifics OSS (MIT) free; pay only Azure/model consumption vendor
- Agent abstraction ReAct + plan-and-execute vendor
- Memory primitives multiple (vector + key-value via plugin) vendor
- LLM lock any (BYO via connectors) vendor
- Runtimes Python, .NET, Java vendor
- Session handling sticky session (ChatHistory per Kernel; ChatHistoryReducer for truncation) vendor
- Validated verticals Microsoft enterprise, .NET / C# / Java agent developers vendor
- Time to running <10min (NuGet install) vendor
- Anti-fit not for non-Microsoft / non-SK stacks vendor
- Optimised for Microsoft enterprise integration (Azure-first) vendor
- Adjacent infrastructure requires Semantic Kernel; BYO LLM (Azure-first); BYO vector store vendor
- Pros Microsoft-backed; first-class .NET support; integrates cleanly with Azure / M365 stacks. vendor
- Cons .NET/Azure-shaped; smaller mind-share among Python-first AI engineers. vendor
- Project & sources GitHub microsoft/semantic-kernel SK agent memory docs vendor
- commit-trajectory [{"ym":"2025-02","cum":4059},{"ym":"2025-03","cum":4214},{"ym":"2025-04","cum":4365},{"y… repo
- download-trajectory [{"ym":"2025-11","monthly":961128,"cum":961128},{"ym":"2025-12","monthly":1935103,"cum":… registry
- cost-input-usd-per-mtok 0 vendor
- cost-output-usd-per-mtok 0 vendor
- cost-tier free vendor
- cost-pricing-model free-tier vendor
- cost-last-verified 2026-06-29 vendor
- architecture-deployment-locus local vendor
- licensing-bucket open-source repo
- positioning-platform-vs-product platform vendor
- taxonomy-three-layer-rollup agent-infra repo
- developer-activity-opendigger {"activity":179.28,"as_of":"2026-09","openrank_12mo":10.92,"openrank_current":6.89,"part… third-party
- sdk-languages csharp,java,python repo
- protocol-support mcp-client repo
- maturity ga registry
What changed
Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)
- 2026-10-23
- Compliance “Inherits Azure compliance posture: SOC 2 Type II, ISO 27001, HIPAA, PCI DSS, GDPR; compliance templates built into …” → “Inherits Azure compliance posture: SOC 2 Type II, ISO 27001, HIPAA, PCI DSS, GDPR” source
- 2026-10-22
- Latest release “dotnet-1.75.0 2026-04-29” → “python-1.45.0 (2026-10-06)” source
Related systems
References (1)
- Azure Machine Learning depends on at runtime — adjacent-infrastructure cell: requires Semantic Kernel; BYO LLM (Azure-first); BYO vector store
Referenced by (1)
- Microsoft Agent Framework succeeds — the successor to Semantic Kernel and AutoGen.