Google ADK Memory

https://google.github.io/adk-docs/sessions/memory/

Google's Agent Development Kit. Session services: InMemory / Database / VertexAI. Long-term: InMemoryMemoryService or VertexAIMemoryBankService (uses Vertex AI Memory Bank Preview for persistent cross-session memory).

At a glance

Type
SessionService tiers + VertexAIMemoryBank
Tier
T2
Created
2025-04 (launched at Google Cloud NEXT, April 9 2025)
Latest release
not applicable — not OSS
License
not applicable — not OSS
GitHub
not applicable — no GitHub repo
Pricing
ADK OSS free; Vertex AI Memory Bank and Agent Engine: consumption-based pay-per-use cloud pricing (serverless); sta…
Funding
Google (subsidiary of Alphabet); no independent funding

Taxonomy

storage
kv
retrieval
injection
persistence
cross-session
update
overwrite
unit
episode
governance
inspectable
conflict
overwrite

When to use

Optimised for: Google Cloud / Vertex AI integration

Anti-fit: not for non-GCP stacks

Pros & cons

Pros

Google's Agent Development Kit with memory built in; integrates cleanly with Vertex AI Memory Bank.

Cons

Google-cloud-tilted; less idiomatic outside GCP.

Claims & capabilities

Two MemoryService implementations: InMemoryMemoryService (keyword search, dev/test) and VertexAiMemoryBankService (managed Google Cloud Memory Bank; default for Agent Platform Runtime). Memory Bank intelligently extracts and stores conversation events as long-term "memories" retrievable by semantic query

Technical surface

API surface
SDK: Python + Java (MemoryService, SessionService interfaces)
Backend storage
pluggable (Vertex AI Memory Bank)
Deployment
Both (ADK OSS self-hosted; one-command deploy to Google Cloud Agent Engine / Cloud Run / GKE)
Embedding model
multiple supported
Multi-tenancy
namespace
MCP
native (first-party) — ADK MCP integration
A2A
supported (native) — ADK is A2A reference impl
OpenTelemetry
first-class — Cloud Trace / OTel

Similar systems

Other framework-embedded memory in the catalog, ranked by inbound references.

  • 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

    ListMemory chronological context + teachable agents that vectorise corrections. Integrates with Mem0/Zep rather than building deep memory natively.

  • CrewAI Memory T2

    Memory subsystem inside the CrewAI orchestration framework; integrates with Mem0 for the long-term tier.

  • AGiXT Adaptive Memory T2

    Open-source AI automation platform. Routes between short-term and long-term memory adaptively across any LLM provider; plugin system for storage backends. Memory managed at the instruction-management layer — task context, instruction state, conversation history as unified agent state.

  • Agno (Phidata) Memory T2

    Agno (formerly Phidata). AgentStorage persists sessions to a DB; AgentMemory auto-classifies/store user preferences and conversation summaries. Single-line integrations with LanceDB, Pinecone, Weaviate, Qdrant.

  • Botpress LLMz T1

    Per-plan vector-DB storage quota + LLMz autonomous engine (in-session working memory) + Knowledge Base (semantic search over uploaded docs). Long-term user memory persists across sessions.

Row last verified 2026-05-14. Catalog data is CC-BY-4.0 — see how to read this.