Agent Memory / Context Management

Tool
Category
Segment
Platform / Tool
Plan / License
Monthly Price USD
Pricing Model
Free Tier / OSS
Included Usage / Limits
Memory Model / Types
Persistence / Retrieval
Context Management / Personalization
Integrations / Frameworks
Deployment / Hosting
Security / Privacy
Team / Governance
Best Fit
Main Limits / Caveats
No tagline
Agent Memory / Context ManagementStateful agent runtimeLetta API / Letta CodeOpen source plus Letta Cloud$20/month API plan; personal Pro $20/month; free BYOK/local pathUsage-based API plan plus personal quota plansFree Letta Code can run with BYOK/local models; free Constellation accounts support up to three agents with managed state; API plan adds $0.10 per active agent/month and $0.00015/sec tool executionMemGPT-style self-editing memory blocks, messages, conversations, MemFS context repositories and sleep-time/reflection conceptsAll agent state including memories, user messages, reasoning and tool calls is persisted in a database; MemFS stores git-backed markdown memoryAgents can edit their own memory and keep core memories in context while retrieving older state after compactionLetta API, Letta Code, MCP tools, server/client tools, OpenAI/Anthropic/OpenRouter/BYOK providersLocal open-source, Docker server, Letta Cloud/Constellation and remote environmentsSelf-hosting supports local control; Cloud features require account; BYOK supportedTeam/Enterprise plans available; Enterprise adds RBAC, SAML/OIDC SSO and dedicated supportStateful agents that need inspectable memory, versioned context and agent self-managementAPI and personal plans are distinct; Docker server can use legacy memory blocks instead of MemFS
No tagline
Agent Memory / Context ManagementTemporal graph memory SaaSZepCommercial cloud SaaS$0 free; Flex $125/month; Flex Plus $375/monthCredit-based plans by episode size; retrieval, storage and users are unmeteredFree plan includes 1,000 credits/month, 2 projects, 5 custom entity/edge types and variable rate limits; Flex includes 50,000 credits/month and 600 RPMTemporal Context Graphs with episodes, entities, facts, relationships, custom entity/edge types and observations on higher plansIngests chat messages, JSON payloads and text into a context graph; retrieval, storage, threads and users cost 0 creditsContext Lake serves long-running agent memory with temporal facts and production context assemblyPython, TypeScript and Go SDKs; Graphiti engine; common agent framework integrationsZep Cloud, Cloud plus BYOK and Enterprise BYOC deployment optionsSOC 2 Type II on paid/cloud paths; Enterprise adds HIPAA BAA, audit logs and longer API log retentionFree/Flex project limits; Enterprise adds custom credits, SLA, unlimited projects and dedicated supportProduction agents needing managed temporal memory and graph-based recall at scaleIngestion credits scale with episode size; free processing priority and rate limits are variable
No tagline
Agent Memory / Context ManagementCloud provider agent memoryAmazon Bedrock AgentCore MemoryAWS managed serviceUsage-basedShort-term events, long-term storage and retrieval pricingNo durable free tier capturedShort-term memory is $0.25 per 1,000 new events; long-term storage is $0.75 per 1,000 records/month with built-in strategies or $0.25 with override/self-managed strategies; retrieval is $0.50 per 1,000 record retrievalsShort-term memory for multi-turn conversations and long-term memory that persists across sessions, with shared memory stores across agentsAgentCore Memory stores events/records and supports retrieval through AWS managed APIsProvides context-aware agents with control over what agents remember and learnLangGraph, LangChain, Strands and LlamaIndexAWS managed AgentCore serviceAWS IAM, regional service controls and AWS content/data policies applyAWS account governance, IAM, CloudWatch observability and service quotasAWS-centered teams building production agents with managed memory and enterprise controlsCharges combine memory plus model and other AgentCore components; availability and regions need confirmation
No tagline
Agent Memory / Context ManagementLearning memory systemHindsight by VectorizeMIT open source plus managed cloud$0 self-hosted; Cloud pay-as-you-goSelf-host free or usage-based token billingSelf-host includes all four memory networks, retain/recall/reflect APIs, MCP server and embedded PostgreSQL; Cloud starts with free creditsFour memory networks with Retain, Recall, Reflect, Iris Extract and Mental Model operationsStores, retrieves and synthesizes memory from agent experience; Cloud provides managed infrastructure and backupsAims at agents that learn from repeated experience instead of only recalling snippetsREST API, Python SDK, MCP, LangChain, CrewAI, LlamaIndex and other integrationsSelf-hosted Docker or Hindsight CloudSelf-host keeps data on customer infrastructure; Enterprise supports BYOC/on-prem and custom SLACloud adds dashboard, usage analytics, team collaboration and support SLA; Enterprise adds SSO/RBACAgents that must retain lessons, failures and behavioral patterns across sessionsCloud costs vary by operation tokens; self-host operations still require model/infrastructure spend
No tagline
Agent Memory / Context ManagementCoding-agent memory MCPOpenMemory by Mem0Mem0 product / MCP memory server$0 software path; Mem0/API costs may applyMemory MCP layer for coding agentsPublic page presents install path and project-scoped memory; exact hosted quotas follow Mem0 account/API setupTyped coding memories such as preferences and implementation details, tagged by project/repoAuto-captures, organizes, searches and injects relevant memory into coding agentsFeeds project-specific context automatically so agents do not need repeated manual promptingMCP-compatible coding agents, IDEs and Mem0 ecosystemLocal/plugin plus Mem0-backed services depending configurationAccess logs show memory added, edited or served; visibility rules and tags are documented on product pageIndividual/project memory management; broader governance follows Mem0 account controlsDevelopers wanting portable memory across Claude Code, Cursor, Codex-like agents and IDE sessionsNot a general backend memory database by itself; exact data path and model/API costs depend on setup
No tagline
Agent Memory / Context ManagementCloud provider sessions and memoryVertex AI Agent Engine Sessions / Memory BankGoogle Cloud pre-GA featureUsage-based; sessions/memory bank billing from 2026-01-28Event-based pricing plus Agent Engine runtime resourcesNo public free tier captured beyond Google account/Express Mode pathsPricing page states billing begins for Code Execution, Sessions and Memory Bank on 2026-01-28; public search result reports Sessions and Memory Bank at $0.25 per 1,000 eventsSessions hold event history and state; Memory is personalized information accessed across multiple sessions for a userSessions maintain chronological events and can be managed via ADK or API callsSupports cross-session continuity and personalization for deployed agentsGoogle ADK, Vertex AI Agent Engine SDK and direct API callsGoogle Cloud Vertex AI Agent EnginePre-GA terms apply; Google Cloud IAM, project and region controls applyProject-level IAM, quotas and billing governanceGoogle Cloud teams deploying ADK/Vertex agents that need managed session state and user memoryPre-GA status means behavior, support and pricing can change; exact regional pricing should be checked in Cloud pricing pages
No tagline
Agent Memory / Context ManagementManaged memory layerMem0 PlatformHosted SaaS plus open-source SDK$0 Hobby; paid from $19/monthPlan-based requests plus custom usage-based pricingHobby includes unlimited end users, 10,000 add requests/month, 1,000 retrieval requests/month and 1 project; Starter is $19/month with 50,000 add and 5,000 retrieval requestsConversation, session, user and organizational memory layers; graph memory and vector-backed memory depending modeManaged platform stores, enriches and retrieves memories with vector store, graph services and rerankersDesigned for persistent personalization so agents remember user and org context without prompt bloatLangChain, CrewAI, Vercel AI SDK, REST API, Python/JS SDKs and 20+ integrationsMem0 Cloud or self-hosted open-source stackDocs warn to avoid storing secrets or unredacted PII; platform docs mention audit logs and workspace governanceProject limits by plan; Enterprise adds unlimited usage, SSO, custom integrations and audit logsTeams needing a production memory API that can be added to agents with minimal codeGraph and advanced governance differ between open-source and platform modes; request limits can be tight on free tier
No tagline
Agent Memory / Context ManagementProvider conversation stateOpenAI Conversations APIOpenAI API featureModel/API usage basedConversation state object used with Responses APIIncluded as API capabilityNo standalone free memory tier captured; Responses/model usage and storage/data-retention policies applyConversation objects store messages, tool calls, tool outputs and other items under a durable identifierSubsequent Responses calls can reference the conversation so the platform prepends stored itemsSimplifies multi-turn state across sessions, devices or jobs without sending full history manuallyOpenAI Responses API, OpenAI SDKs and custom appsHosted OpenAI APIOpenAI API data handling and organization settings apply; zero data retention changes stateful optionsOrganization/project/API key controls and application-managed deletion/retention patternsApps that need hosted conversation continuity but not a full semantic memory layerConversation history is not the same as curated long-term memory; apps still need compaction, retention and relevance policies
No tagline
Agent Memory / Context ManagementAgent framework memory serviceGoogle ADK MemoryOpen-source framework feature$0 SDK; backend/model costs separateMemory service interface with local and Vertex backendsNo ADK software cap; examples can use local memory or VertexAiMemoryBankService with Google Cloud billingSession memory and long-term memory via Memory Bank; memories can be saved from sessions and loaded into agentsADK server connects to a memory service URI; callbacks can auto-save sessions to memoryGives ADK agents cross-session recall and personalization when paired with Memory BankGoogle ADK, VertexAiMemoryBankService, ADK web/api_serverLocal ADK app or Vertex AI Agent EngineData path depends on selected memory service; Vertex path follows Google Cloud termsGovernance follows local deployment or Google Cloud project controlsTeams using Google ADK who want memory without adopting a separate memory vendorMemory quality depends on configured service and callbacks; managed features are tied to Vertex AI maturity and billing
No tagline
Agent Memory / Context ManagementAgent memory and RAGMicrosoft AutoGen MemoryOpen-source framework feature$0 software; storage/model costs separateFramework memory abstraction with vector memory examplesNo software cap; ChromaDB extra and embedding models are needed for vector memory examplesMemory and RAG stores, including ChromaDBVectorMemory for semantic retrievalMemory stores content and updates agent context with relevant retrieved itemsAdds persistent context or RAG memory to AutoGen agents during conversationsAutoGen AgentChat, autogen-ext memory components, ChromaDB and custom Memory implementationsApplication code / self-hostedData path depends on chosen vector DB and embedding providerNo hosted governance in OSS; app owns retention and tenant isolationAutoGen users adding retrieval-backed memory to agentsReference implementation is vector/RAG-oriented; advanced memory policies must be built around it
No tagline
Agent Memory / Context ManagementPersonal agent memory hubMembaseCommercial SaaS$0 Free; Pro $20/monthPlan-based personal memory hubFree includes limited memory searches, episodes, wiki documents and AI chats; Pro adds unlimited memory searches and larger limitsSelf-evolving personal memory with episodes, wiki documents, AI chats and MCP integrationAutomatically builds personal memory from daily agents and appsKeeps context across Cursor, ChatGPT, Claude Desktop, Claude Code, Codex, VS Code and MCP-compatible appsMCP-compatible agents, Gmail, Calendar, Slack, ChatGPT/Claude/Gemini importers and coding toolsHosted SaaSPricing page states opt out of model training and account deletion controls; full security details need account reviewFree/Pro personal plan controls; community/priority support by planPower users wanting one personal memory across many AI toolsLess developer-backend-oriented than Mem0/Zep; limited public details on enterprise controls
No tagline
Agent Memory / Context ManagementAgent framework memoryAgno MemoryOpen-source framework feature$0 software; database/model costs separateDatabase-backed framework memoryNo software cap; memories are stored in the connected database such as SQLite, Postgres or MongoDBAutomatic memory and agentic memory; chat history, user preferences and session summaries in v1 docsStores memories in agent database tables/collections and retrieves them on future runsupdate_memory_on_run automatically extracts memories; enable_agentic_memory gives the agent tools to manage memoryAgno agents, teams, SQLite, Postgres, MongoDB and major LLM providersApplication code / self-hostedData stays in configured database; model provider choice controls memory extraction data pathDatabase and application controls; no standalone governance layer in core memory featureAgno users who need simple persistent user memory in agent appsAutomatic and agentic memory modes are mutually exclusive; memory quality depends on LLM extraction
No tagline
Agent Memory / Context ManagementOpen-source memory frameworkMemoryScopeOpen-source / token ecosystemPublic fixed price not capturedPackage plus hosted ecosystemPublic site links documentation and GitHub; exact free quota or paid pricing was not capturedLong-term memory capabilities for LLM chatbots and agents; collective experience in a MemoryScopePackage API creates/manages agents and interacts with their memoriesAims to give agents flexible long-term memory and shared experiencePython package, GitHub project and MemoryScope ecosystemPackage/local plus hosted ecosystem signalsSecurity/privacy details were not publicly itemized in captured sourceGovernance details not capturedResearchers/devs exploring a dedicated long-term memory framework from local seedsOfficial pricing, hosting and enterprise controls are unclear; verify project maturity before relying on it
No tagline
Agent Memory / Context ManagementAI-native memory libraryHonchoOpen source plus managed servicePublic fixed price not capturedMemory library with managed serviceDocs describe open-source memory library and managed service; public pricing was not found in official docs searchLong-term memory about entities such as users, agents, groups and ideas; social intelligence over stored historyStores history and provides tools to traverse it and infer latent contextBuilds state about changing entities over time for personalized and social agentsAny model/framework/architecture through Honcho APIs/librarySelf-hosted/open source or Honcho managed serviceData/privacy details require review of Honcho deployment docs and termsManaged service governance not publicly priced in captured docsAgents requiring rich identity, social context and entity-centered memoryPricing and quotas are not clearly public; evaluate maturity and hosting model before production
No tagline
Agent Memory / Context ManagementOpen-source cognitive memoryEngramApache-2.0 open source$0 self-hosted; planned Developer $29/month; planned Team $149/monthSelf-host free; managed cloud in developmentSelf-hosted plan lists unlimited agents and memories with all 4 cognitive memory types; Developer and Team cloud plans are marked coming soonSemantic, episodic, procedural and working memory with confidence scoring, decay, contradiction detection and lifecycle managementREST API stores and recalls typed memories with hybrid vector plus graph retrievalHot memories can auto-inject into prompts while cold memories surface only when relevantPython SDK, REST API, LangChain integration, multi-LLM support and Docker ComposeSelf-hosted Go/Postgres/pgvector; cloud waitlistSelf-hosting owns data completely; managed cloud not yet generally availableCommunity support in self-hosted; planned Team includes workspaces, dashboard and SLATeams wanting inspectable open-source cognitive memory with decay and contradiction handlingCloud plans are not live; product/domain variants named Engram may be confusing and need validation
No tagline
Agent Memory / Context ManagementEdge platform agent memoryCloudflare Agents MemoryCloudflare platform featureWorkers/Durable Objects pricing appliesPlatform memory feature for Cloudflare AgentsIncluded with platform capabilityNo separate memory price captured; Cloudflare Workers, storage and related platform billing applyConversation history plus context memory blocks that can be read, written, searched and loadedCloudflare Agents manage persisted session state and context memory blocksKeeps agent conversations and reusable context available inside Cloudflare edge-hosted agentsCloudflare Agents SDK, Workers, Durable Objects and platform storageCloudflare edge platformCloudflare account/security controls and platform data policies applyCloudflare project/team controls and deployment governanceDevelopers building agent apps directly on Cloudflare's platformPlatform-specific; not a vendor-neutral memory layer and pricing depends on underlying Cloudflare resources
No tagline
Agent Memory / Context ManagementCloud provider long-term memoryMicrosoft Foundry Agent Service MemoryAzure public previewUnderlying model usage; memory pricing may changePreview memory with underlying chat and embedding model billingNo separate free tier capturedDocs list maximum 100 scopes per memory store, 10,000 memories per scope, 1,000 search requests/min and 1,000 update requests/minLong-term memory with user profile memory and chat summary memory; stores durable memory items in a managed memory storeExtraction, consolidation and retrieval phases manage preferences, summaries and conflicting factsDesigned for continuity across sessions, devices and workflows with personalized responsesMicrosoft Foundry Agent Service, Memory Store API, Azure OpenAI chat and embedding deploymentsAzure managed servicePreview terms apply; requires compatible Azure OpenAI chat and embedding model deploymentsAzure subscription, scopes, IAM/project controls and Foundry Agent Service governanceAzure teams needing a managed long-term memory store tied to Foundry agentsPreview feature; automatic scope resolution is limited and pricing/billing can change during preview
No tagline
Agent Memory / Context ManagementAgent memory componentLlamaIndex MemoryMIT open-source framework feature$0 software; model/storage costs separateFramework memory abstractionNo software cap; memory implementations depend on selected stores, LLMs and embeddingsShort-term FIFO message memory with optional long-term extracted memory; BaseMemory can be customizedAgent calls memory put/get or add/get; memory can trim to context window and optionally persist extracted informationKeeps agents within context limits and can add user/static blocks or long-term memory sourcesLlamaIndex Python, LlamaIndexTS, agents, chat engines, memory blocks and adaptersEmbedded in application codeSecurity depends on chosen storage, models and app controlsNo separate team governance in the componentExisting LlamaIndex apps needing memory that fits the agent/query-engine ecosystemSome older memory classes are deprecated or being replaced; production durability is not automatic
No tagline
Agent Memory / Context ManagementLong-term memory SDKLangMemMIT$0 softwareOpen-source SDK; model, embedding and storage costs separateNo software cap; production persistence requires a DB-backed LangGraph store such as AsyncPostgresStoreStructured long-term memories, hot-path memory tools and background memory managersExtracts, consolidates, updates and searches memories using LangGraph BaseStore-compatible storageAgents can manage memory during conversations or background processes can update it after interactionsLangGraph, LangChain, Python package, any BaseStore-compatible storageApplication code / self-hostedData path depends on model and storage providers; local stores can keep data localNo hosted team governance in the SDK itselfLangGraph teams needing first-party long-term memory primitivesPython-centric; in-memory examples are not durable; memory extraction can add LLM latency/cost
No tagline
Agent Memory / Context ManagementMemory management kitReMeOpen-source GitHub project$0 softwareOpen-source memory management kit; model/storage costs separateNo software cap captured in repository listingPersonal memory, task memory and tool memory for agentsFramework manages memory refinement and retrieval for agent workflowsHelps agents remember, refine and reuse information across interactionsAgentScope/modelscope ecosystem and Python agent workflowsSelf-hosted/local application codeData path depends on configured models and storageNo hosted governance capturedDevelopers who want a research/open-source kit for structured agent memory typesOfficial docs/pricing are GitHub-centric; production support and governance are limited
No tagline
Agent Memory / Context ManagementLocal-first shared memoryMemobaseLocal-first plus cloud sync$0 Free; Pro $9/month; Unlimited $29/monthCredit-based monthly plansFree includes 500 credits/month; store_memory costs 10 credits and search costs 1 credit; Pro includes 5,000 credits/month; Unlimited lists unlimited creditsPersistent shared memory for AI agents with semantic vector search, per-user isolation and local SQLite modeLocal mode stores memories, entity relationships and architectural context in SQLite; cloud sync available with loginAdds cross-tool context continuity for Claude, ChatGPT, Claude Code and custom MCP agentsCLI, MCP server, Claude, Claude Desktop, ChatGPT and custom connector/API paths100% private local mode or managed cloud syncLocal mode sends no data to Memobase servers; cloud sync requires service trustFree/Pro/Unlimited plan controls; dedicated support and SLA on UnlimitedIndividuals and small teams wanting local-first shared memory for coding and chat agentsTwo Memobase product lineages exist on the web; verify domain/product before adoption
No tagline
Agent Memory / Context ManagementRedis-backed memory serviceRedis Agent MemoryOpen source plus Redis Cloud service$0 OSS; Redis Cloud pricing appliesSelf-host Redis service or Redis Cloud private previewOSS server has no software cap; Redis Cloud service availability/pricing depends on account and preview accessTwo-tier memory: session/working memory with TTL and long-term persistent memoryREST API and client libraries store, retrieve and manage contextual data in Redis with vector/keyword/hybrid searchAutomatic lifecycle management, schemas and TTL keep agent memory structured and boundedREST API, MCP, Python SDK, OpenAI, Anthropic and 100+ providers through LiteLLMSelf-hosted Redis Agent Memory Server or Redis CloudSelf-hosting keeps data in customer Redis; Redis Cloud controls depend on planAPI key management and configurable schemas; broader governance follows Redis deploymentTeams already using Redis that want a production-ready memory service rather than custom tablesManaged cloud path is private preview; requires Redis/vector setup and model API keys for extraction
No tagline
Agent Memory / Context ManagementFramework persistence and memoryLangGraph MemoryOpen-source framework feature / LangGraph Platform$0 OSS; platform/infrastructure costs separateApplication-managed stores and checkpointersNo OSS software cap; in-memory stores are ephemeral and production needs Postgres or another persistent backendShort-term thread-level persistence and long-term user/application-level memoryCheckpointers save graph state; stores hold memories across threads and conversationsManages conversation history, trimming and cross-session memories inside LangGraph agentsLangGraph, LangChain, Postgres stores, InMemoryStore and LangGraph PlatformSelf-hosted app, LangGraph Platform or customer infrastructureSecurity depends on selected database and platform; application owns scoping and retentionGovernance depends on deployment and LangSmith/LangGraph Platform if usedLangGraph apps needing stateful workflows and memory without a separate SaaSInMemorySaver loses data on restart; long-term memory design and scoping are still app responsibilities
No tagline
Agent Memory / Context ManagementAgent framework memory providerSemantic Kernel Agent MemoryExperimental open-source framework feature$0 software; Mem0/model/storage costs separateAgentThread memory components and provider integrationsNo separate software cap; the documented Mem0Provider uses the Mem0 service and its pricing/quotasMemories extracted from thread messages and supplied to agents; Mem0Provider enables cross-thread user memoryAgentThread components capture, retain and surface memory as neededAllows agents to remember user preferences and context across multiple threadsSemantic Kernel Agents, Microsoft.SemanticKernel.Memory.Mem0Provider and vector storesApplication code / self-hosted with optional external memory serviceDocs mark feature experimental; data path depends on Mem0 or configured memory providerGovernance follows app, provider and Azure/Microsoft stack if usedSemantic Kernel teams needing a Microsoft-native way to plug memory into AgentThread flowsExperimental and subject to change; current official guide centers on Mem0 integration rather than a standalone managed memory store
No tagline
Agent Memory / Context ManagementAgent SDK session memoryOpenAI Agents SDK SessionsOpen-source SDK feature$0 SDK; API/model/storage costs separateBuilt-in session memory with pluggable storageSDK has no software cap; local SQLite or hosted OpenAI Conversations storage can be used depending session implementationSession memory stores conversation history for a specific session; sandbox agent memory is separate for reusable tips/preferencesSessions automatically maintain conversation history across multiple agent runsEliminates manual to_input_list chaining and supports trimming/compaction patterns in SDK guidesOpenAI Agents SDK Python/JS, OpenAI ConversationsSession, SQLite/custom sessionsApplication code plus optional OpenAI-hosted storageCustom sessions can enforce retention policies, encryption and metadata; hosted storage follows OpenAI API policiesGovernance depends on app storage implementation and OpenAI org controlsDevelopers building OpenAI Agents SDK apps that need short-term continuity and pluggable persistenceSession memory is primarily conversation history, not autonomous semantic/episodic memory unless paired with other tools