Pydantic-AI MemoryTool
https://pydantic.dev/docs/ai/tools-toolsets/builtin-tools/
Memory exposed via Anthropic SDK's BetaAbstractMemoryTool when Anthropic provider is active. No built-in persistent storage in the framework itself. Other providers require external memory (Hindsight, Mem0) wired in manually.
At a glance
- Type
- Tool-delegated to model provider
- Tier
- T3
- Section
- Framework-embedded memory
- Created
- 2024-12 (pydantic-ai launched December 2024)
- Latest release
- not applicable — not OSS
- License
- not applicable — not OSS
- GitHub
- not applicable — no GitHub repo
- Pricing
- OSS free (MIT for pydantic-ai); pay for model provider API usage; Pydantic Logfire observability has separate pricing
- Funding
- Pydantic (company) raised $17.2M total; Series A $12.5M Oct 2024; investors include Sequoia Capital and Partech Par…
Taxonomy
- storage
- proprietary
- retrieval
- agentic
- persistence
- cross-session
- update
- agent-controlled
- unit
- fact
- governance
- opaque
- conflict
- none
When to use
Optimised for: model-provider memory delegation (DX)
Anti-fit: not for non-Anthropic providers (delegates to Anthropic memory tool)
Pros & cons
Pros
Type-safe memory primitive inside Pydantic-AI's agent design; clean SDK ergonomics.
Cons
Thinner feature surface than dedicated memory layers.
Claims & capabilities
Memory capability entirely delegated to model provider.
Technical surface
- API surface
- not applicable — research paper
- Backend storage
- not applicable — research paper
- Deployment
- Both (OSS self-hosted; Pydantic Logfire cloud for observability)
- Embedding model
- not applicable — research paper
- Multi-tenancy
- not applicable — research paper
- MCP
- native (first-party)
- A2A
- supported — Pydantic-AI A2A
- OpenTelemetry
- first-class — Logfire
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.
Related systems
References (1)
- Anthropic Claude (foundation models) depends on at runtime — Memory exposed via Anthropic SDK's BetaAbstractMemoryTool when Anthropic provider is active.