AutoGen Memory
https://microsoft.github.io/autogen/
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.
Maintenance · Verified 2026-10-23 · fresh as of 2026-10-23 · 62 cited facts, 7 with a third-party source · 2 recorded changes · Suggest a correction
At a glance
- Type
- List + teachable vectors vendor fix
- Tier
- T2
- Section
- Framework-embedded memory
- Latest release
- python-v0.7.5 (2025-09-30) — unchanged; no newer release registry fix
- Pricing
- OSS (MIT); no direct cost for library repo fix
- Funding
- Microsoft-backed; AutoGen is an open-source Microsoft Research project with no independent funding third-party fix
- Freshness
- 2026-10-23 fix
- Layer
- agent-infra repo fix
- Licensing
- open-source repo fix
- Protocols
- mcp-client repo fix
- Developer activity
- OpenRank 7.63 (12-mo avg 6.48) · 108 participants — 2026-09 third-party fix
Memory taxonomy
- storage
- kv
- retrieval
- injection
- persistence
- cross-session
- update
- append-only
- unit
- episode
- governance
- inspectable
- conflict
- append
When to use
Optimised for: Microsoft Research multi-agent + teachable corrections
Anti-fit: not for non-AutoGen stacks
Pros & cons
Pros
Microsoft Research origin; strong multi-agent conversation memory model.
Cons
AutoGen's broader feature drift over time has affected memory abstractions; less stable than dedicated layers.
Claims & capabilities
Framework adoption nearly doubled YoY: ~9% of organizations early-2025 to ~18% early-2026 per Datadog State of AI Engineering. AutoGen now in maintenance mode — superseded by Microsoft Agent Framework. AutoGen Bench provides agent benchmark suite; matches LangGraph in token use and latency
Technical surface
- API surface
- SDK: Python, .NET
- Backend storage
- pluggable
- Deployment
- Both (OSS self-hosted Python; Azure AI Foundry managed runtime with GA Q1 2026)
- Embedding model
- multiple supported
- Multi-tenancy
- namespace
- MCP
- via official adapter — autogen-ext-mcp
- A2A
- searched not found
- OpenTelemetry
- first-class — AutoGen has OTel runtime
Alternatives to AutoGen 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.
- 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.
- Semantic Kernel Memory T2
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.
Sources (61 cited facts, 7 third-party)
Each fact links to the page it came from; the badge says who published it (how to read sources).
- What it is List + teachable vectors vendor
- What it does & what's distinct AutoGen's memory layer: ListMemory chronological context plus teachable agents that vect… repo
- Claims / benefits Framework adoption nearly doubled YoY: ~9% of organizations early-2025 to ~18% early-202… third-party
- Modalities text repo
- Created 2023-08 vendor
- Latest release python-v0.7.5 (2025-09-30) — unchanged; no newer release registry
- License MIT repo
- GitHub microsoft/autogen — ~60.4k stars, maintenance mode repo
- Mindshare 1 inbound vendor
- Funding Microsoft-backed; AutoGen is an open-source Microsoft Research project with no independe… third-party
- Customers / scale 10,000+ organizations using Azure AI Foundry Agent Service (Microsoft umbrella); no Auto… third-party
- Pricing OSS (MIT); no direct cost for library repo
- Compliance Inherits Microsoft/Azure compliance posture (SOC 2, ISO 27001, HIPAA, GDPR etc.) when de… third-party
- Data handling Inherits Azure data handling when deployed on Azure; OSS self-hosted: data handling left… repo
- Deployment Both (OSS self-hosted Python; Azure AI Foundry managed runtime with GA Q1 2026) third-party
- HQ US (Redmond, WA — Microsoft HQ) third-party
- Founders / pedigree Microsoft Research; open-source multi-agent framework from Microsoft vendor
- Memory volume / scale no scale claim repo
- API surface SDK: Python, .NET vendor
- Backend storage pluggable vendor
- Multi-tenancy namespace vendor
- Embedding model multiple supported vendor
- Versioning immutable append-log vendor
- Tombstoning append-only no delete vendor
- Namespace primitives per-tenant vendor
- Contradiction handling always append (no resolution) repo
- Forgetting policy none (permanent) vendor
- MCP support via official adapter — autogen-ext-mcp vendor
- OTel first-class — AutoGen has OTel runtime vendor
- Import / export JSON state save/load vendor
- Integration count 1 repo
- Orchestration multi-agent supervisor vendor
- Programmatic control full CRUD via Python SDK vendor
- Pricing specifics OSS (MIT) free; Azure AI Foundry consumption-based when managed repo
- Agent abstraction multi-agent supervisor (group chat) vendor
- Memory primitives multiple (ListMemory chronological + teachable vector) vendor
- LLM lock any (BYO via OpenAI-compatible adapters) vendor
- Runtimes Python repo
- Session handling sticky session (conversable agent) vendor
- Validated verticals research, agent developers (horizontal) vendor
- Time to running <10min (pip install autogen) vendor
- Anti-fit not for non-AutoGen stacks vendor
- Optimised for Microsoft Research multi-agent + teachable corrections vendor
- Adjacent infrastructure requires AutoGen; BYO LLM + vector store vendor
- Pros Microsoft Research origin; strong multi-agent conversation memory model. vendor
- Cons AutoGen's broader feature drift over time has affected memory abstractions; less stable … vendor
- Project & sources GitHub microsoft/autogen vendor
- commit-trajectory [{"ym":"2024-11","cum":2803},{"ym":"2024-12","cum":2958},{"ym":"2025-01","cum":3196},{"y… repo
- download-trajectory [{"ym":"2025-11","monthly":81884,"cum":81884},{"ym":"2025-12","monthly":162109,"cum":243… registry
- cost-input-usd-per-mtok 0 repo
- cost-output-usd-per-mtok 0 repo
- cost-tier free repo
- cost-pricing-model free-tier repo
- cost-last-verified 2026-06-29 repo
- 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":193.81,"as_of":"2026-09","openrank_12mo":6.48,"openrank_current":7.63,"parti… third-party
- sdk-languages python repo
- protocol-support mcp-client repo
What changed
Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)
- 2026-10-23
- Pricing “OSS (MIT); no direct cost for library; Azure AI Foundry Agent Service for managed hosting uses Azure consumption pr…” → “OSS (MIT); no direct cost for library” source
- Funding / ownership “Microsoft-backed; AutoGen is an open-source Microsoft Research project with no independent funding; merged into Mic…” → “Microsoft-backed; AutoGen is an open-source Microsoft Research project with no independent funding” source
- 2026-10-22
- Latest release “python-v0.7.5 2025-09-30” → “python-v0.7.5 (2025-09-30) — unchanged; no newer release” source
Related systems
References (3)
- Chroma integrates with — Microsoft AutoGen autogen-ext ships a ChromaDB vector memory implementation (autogen_ext.memory.chromadb). (verified 2026-10-29)
- Mem0 integrates with — ListMemory chronological context + teachable agents that vectorise corrections. Integrates with Mem0/Zep rather than building deep memory natively.
- Zep & Graphiti integrates with — Integrates with Mem0/Zep rather than building deep memory natively.
Referenced by (2)
- mcp-memory-service (doobidoo) integrates with — Agent-pipeline focused (LangGraph, CrewAI, AutoGen)
- Zep & Graphiti integrates with — Zep docs have a Microsoft AutoGen integration page (zep-autogen package). (verified 2026-10-29)