# Memory layers for AI agents > 8 agent memory listings ranked by the Anchor benchmark. Leader Zep (B). Services that store what an agent learns about a user or a task and bring back the right parts later: facts, preferences, conversation summaries and knowledge graphs. Compared on recall quality, latency, what they keep and how you delete it. - Canonical: https://www.anchorterminal.com/categories/agent-memory - Markdown: https://www.anchorterminal.com/categories/agent-memory.md (~5,200 tokens) - Slim: https://www.anchorterminal.com/categories/agent-memory.min.md (~480 tokens, same facts, less prose, for token-sensitive contexts) - JSON: https://www.anchorterminal.com/categories/agent-memory.json (this page as data, same URL with Accept: application/json) - Site index for agents: https://www.anchorterminal.com/llms.txt (full text: https://www.anchorterminal.com/llms-full.txt) - API: https://www.anchorterminal.com/api/v1/index.json - Updated: 2026-10-05 Services that store what an agent learns about a user or a task and bring back the right parts later: facts, preferences, conversation summaries and knowledge graphs. Compared on recall quality, latency, what they keep and how you delete it. - Tools ranked: 8 · agent-ready (BB or better): 0 · accept x402: 1 · hosted endpoints: 6 · desk reviews by the panel: 16 - JSON: https://www.anchorterminal.com/api/v1/tools.json (list) · https://www.anchorterminal.com/api/v1/rankings.json (ranked) · https://www.anchorterminal.com/api/v1/x402.json (payable) · https://www.anchorterminal.com/api/v1/capabilities.json (by capability) - Grades run AA, A, BB, B, C, D, E, F · methodology: https://www.anchorterminal.com/benchmark/ - Capabilities in this category: memory.store, memory.search, memory.graph, memory.user, memory.delete - https://letme.dev/memory.store picks the top-graded tool in this list and says how to call it direct; calling through letme comes later (https://www.anchorterminal.com/letme/index.md) ## Ranking | # | Tool | Vendor | Kind | Category | Grade | Score | Confidence | x402 | Auth | Where | Reviews | Page | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | 112 | Zep | Zep | HTTP API | Memory | B | 69.6 | high | no | OAuth or key | hosted | 4/5 (2) | https://www.anchorterminal.com/tools/zep.md | | 188 | Honcho | Plastic Labs | HTTP API | Memory | B | 64.2 | medium | yes ($0.001/call) | OAuth or key | hosted | 3/5 (2) | https://www.anchorterminal.com/tools/honcho.md | | 201 | Supermemory API + MCP | Supermemory | HTTP API | Memory | B | 63.6 | medium | no | OAuth or key | hosted | 3.5/5 (2) | https://www.anchorterminal.com/tools/supermemory.md | | 301 | Mem0 Platform + MCP | Mem0 | HTTP API | Memory | C | 56.6 | medium | no | API key | hosted | 2.5/5 (2) | https://www.anchorterminal.com/tools/mem0.md | | 320 | Cognee | Cognee | Model platform | Memory | C | 54.6 | medium | no | OAuth or key | hosted + local | 3/5 (2) | https://www.anchorterminal.com/tools/cognee.md | | 322 | Memory (MCP reference server) | MCP project (reference servers) | MCP server | Databases | C | 54.4 | medium | no | None | local | 2.5/5 (2) | https://www.anchorterminal.com/tools/memory-reference-server.md | | 333 | Graphiti | Zep | Agent framework | Memory | D | 53.3 | medium | no | None | library | 2.5/5 (2) | https://www.anchorterminal.com/tools/graphiti.md | | 359 | Hindsight | Vectorize | HTTP API | Memory | D | 50.4 | medium | no | OAuth or key | hosted | 3/5 (2) | https://www.anchorterminal.com/tools/hindsight.md | Scores are from public evidence against the published checklist (https://www.anchorterminal.com/benchmark/), with Performance and Task success pending. p95 latency and context cost come from our probes, which haven't run yet. ## Summaries ### 112. Zep, B (69.6) Hosted context and memory service built on a temporal graph of facts, relationships and source episodes. Facts that new data invalidates keep the time they stopped being true. The terms of 17 August 2026 grant a perpetual, irrevocable licence to train models on customer data. - Page: https://www.anchorterminal.com/tools/zep · Markdown: https://www.anchorterminal.com/tools/zep.md · JSON: https://www.anchorterminal.com/api/v1/tools/zep.json - Capabilities: memory.store, memory.search, memory.graph, memory.user, memory.delete · endpoint: `https://api.getzep.com/api/v2` ### 188. Honcho, B (64.2) Memory API that models each participant (a peer) in a conversation. Pay per call over x402 or MPP at agentcash.honcho.dev, with free read endpoints. No published rate limits, 429 guidance or SLA. - Page: https://www.anchorterminal.com/tools/honcho · Markdown: https://www.anchorterminal.com/tools/honcho.md · JSON: https://www.anchorterminal.com/api/v1/tools/honcho.json - Capabilities: memory.store, memory.search, memory.user, memory.delete · endpoint: `https://api.honcho.dev/v3` ### 201. Supermemory API + MCP, B (63.6) Memory and context API that ingests text, URLs, PDFs, images and video, extracts memories into a graph per container tag (usually one per user) and returns them through search or a user profile endpoint. 8-tool hosted MCP with OAuth and per-space read or write permission. No legal entity named in the terms or privacy policy. - Page: https://www.anchorterminal.com/tools/supermemory · Markdown: https://www.anchorterminal.com/tools/supermemory.md · JSON: https://www.anchorterminal.com/api/v1/tools/supermemory.json - Capabilities: memory.store, memory.search, memory.graph, memory.user, memory.delete · endpoint: `https://api.supermemory.ai` ### 301. Mem0 Platform + MCP, C (56.6) Hosted memory layer that extracts facts from conversations and returns the relevant ones for a user, agent or run on later turns. An agent can create its own Free account with `mem0 init --agent`, no email and no card. Free Plan data is used to train Mem0's models, per the privacy policy of 22 August 2026. - Page: https://www.anchorterminal.com/tools/mem0 · Markdown: https://www.anchorterminal.com/tools/mem0.md · JSON: https://www.anchorterminal.com/api/v1/tools/mem0.json - Capabilities: memory.store, memory.search, memory.graph, memory.user, memory.delete · endpoint: `https://api.mem0.ai` ### 320. Cognee, C (54.6) Open-source memory engine that turns documents, conversations and synced sources into a knowledge graph plus a vector index and answers queries over both. Apache-2.0 library, REST server and MCP server, all self-hostable, with local models and no LLM key since 1.6.0. No status page, rate limits, SLA or SOC 2 for Cloud, and Cloud runs in AWS us-east-1 with no EU region. - Page: https://www.anchorterminal.com/tools/cognee · Markdown: https://www.anchorterminal.com/tools/cognee.md · JSON: https://www.anchorterminal.com/api/v1/tools/cognee.json - Capabilities: memory.store, memory.search, memory.graph, memory.delete · endpoint: `https://.aws.cognee.ai/api/v1` ### 322. Memory (MCP reference server), C (54.4) Knowledge-graph persistent memory reference server (entities, relations, observations) stored as JSONL at MEMORY_FILE_PATH. Nine tools with typed schemas and accurate read, destructive and idempotent annotations. No pagination or limits. `read_graph` returns everything and `search_nodes` every match. - Page: https://www.anchorterminal.com/tools/memory-reference-server · Markdown: https://www.anchorterminal.com/tools/memory-reference-server.md · JSON: https://www.anchorterminal.com/api/v1/tools/memory-reference-server.json - Capabilities: memory.graph ### 333. Graphiti, D (53.3) Open-source Python framework from Zep that builds a temporal knowledge graph from chat messages, text and JSON. Apache-2.0, with FalkorDB, Neo4j or Amazon Neptune as the store. Every add runs LLM extraction, so ingestion costs tokens and time. - Page: https://www.anchorterminal.com/tools/graphiti · Markdown: https://www.anchorterminal.com/tools/graphiti.md · JSON: https://www.anchorterminal.com/api/v1/tools/graphiti.json - Capabilities: memory.store, memory.search, memory.graph, memory.delete ### 359. Hindsight, D (50.4) Memory engine for storing and retrieving information used by agents. API keys support bank-level restrictions, expiry and child-key revocation. Retain ingestion costs $10 per million tokens. - Page: https://www.anchorterminal.com/tools/hindsight · Markdown: https://www.anchorterminal.com/tools/hindsight.md · JSON: https://www.anchorterminal.com/api/v1/tools/hindsight.json - Capabilities: memory.store, memory.search, memory.delete · endpoint: `https://api.hindsight.vectorize.io` ## How we test this category One user's history fed in over several sessions, then questions that need facts from early on, a changed preference and a deleted fact. We check what comes back, how fast, and whether the deletion sticks. This test hasn't run yet, so Task success is pending and the grades here come from the categories assessed from public evidence. ## Indexed, not reviewed (63) Sorted into this category from public catalogues, with facts and our own checks but no score, grade or rank (https://www.anchorterminal.com/indexed/index.md). | Listing | Kind | What it does | Why it's here | | --- | --- | --- | --- | | [6DuckLearn MCP](https://www.anchorterminal.com/tools/6ducklearn-mcp.md) | MCP server | Connect agents to 6DuckLearn memory, approvals, and runtime control. | vendor's own | | [Achriom](https://www.anchorterminal.com/tools/achriom.md) | MCP server | Media memory for AI agents and their humans: books, movies, music, shows, anime, podcasts, games. | vendor's own | | [adrkit decision memory](https://www.anchorterminal.com/tools/adrkit-mcp.md) | MCP server | Deterministic, offline, read-only ADR decision memory for coding agents. No model or network calls. | vendor's own | | [AI Context Flow](https://www.anchorterminal.com/tools/plurality-ai-context-flow.md) | MCP server | Universal memory for AI agents and tools. Save, organize and search context anywhere. | vendor's own | | [Amber](https://www.anchorterminal.com/tools/ambermem-amber.md) | MCP server | Long-term memory for AI assistants. Hybrid retrieval, query expansion, auto-topics. | vendor's own | | [Andish](https://www.anchorterminal.com/tools/andish-mcp.md) | MCP server | Knowledge-graph memory for your team's documents. Answers cite the source, not a summary. | vendor's own | | [being-mcp-server](https://www.anchorterminal.com/tools/ruddia-being-mcp-server.md) | MCP server | Personality Runtime for AI agents — persistent memory, personality, and relationships via MCP | vendor's own | | [Bluet](https://www.anchorterminal.com/tools/bluet.md) | MCP server | Product memory your agents read before they work: briefs, plan checks against decisions, debriefs. | vendor's own, widely used | | [Bourdon](https://www.anchorterminal.com/tools/bourdon.md) | MCP server | Recognition-first cross-agent memory federation. One shared memory across all your agents. | vendor's own | | [br8n.io MCP server](https://www.anchorterminal.com/tools/br8n-mcp.md) | MCP server | Owned, portable working memory: a plain-files markdown brain any model can read and cite. | vendor's own | | [Brainfeather](https://www.anchorterminal.com/tools/brainfeather-mcp.md) | MCP server | Long-term memory for AI coding agents: durable project facts, recalled by every MCP client. | vendor's own | | [Cairn.ink Memory](https://www.anchorterminal.com/tools/cairn-memory.md) | MCP server | Cross-session memory for AI agents with a Source Receipt for every memory, over MCP. | vendor's own | | [Citadel](https://www.anchorterminal.com/tools/citadeldb-mcp.md) | MCP server | Encrypted-first embedded database with vector search and agent memory, exposed as MCP tools | vendor's own | | [cluster-mcp](https://www.anchorterminal.com/tools/clusteragent-cluster-mcp.md) | MCP server | Self-hostable index of AI financial agents — tokenized stocks + crypto, memory, LLM gateway. | vendor's own | | [Compose MCP](https://www.anchorterminal.com/tools/smartmemory-compose-mcp.md) | MCP server | Typed feature management for compose projects: roadmap, changelog, artifacts, journal. | vendor's own, widely used | | [ContextStream](https://www.anchorterminal.com/tools/contextstream-mcp.md) | MCP server | Persistent memory and cross-session learning for AI coding assistants (hosted remote MCP). | vendor's own | | [deerdawn](https://www.anchorterminal.com/tools/deerdawn.md) | MCP server | AI session memory: the brief your AI reads before every session so no session starts cold. | vendor's own | | [Docmancer](https://www.anchorterminal.com/tools/docmancer.md) | MCP server | Local-only MCP server for source-attributed Markdown memory and documentation retrieval. | vendor's own | | [Engram](https://www.anchorterminal.com/tools/engram.md) | MCP server | Memory for AI agent teams across tools, sessions, repositories, and teammates. | vendor's own | | [Engram](https://www.anchorterminal.com/tools/getengram-engram.md) | MCP server | Persistent, verbatim, searchable memory for AI assistants — one memory across every MCP client. | vendor's own | | [engrava](https://www.anchorterminal.com/tools/sovantica-engrava.md) | MCP server | MindQL queries, hybrid search, and a thought graph - read/write memory for AI agents. | vendor's own | | [gateway](https://www.anchorterminal.com/tools/wingmanprotocol-gateway.md) | MCP server | Durable self for AI agents: one-call resume, memory, real browser, free chat + hire real humans. | vendor's own | | [gjalla](https://www.anchorterminal.com/tools/gjalla-mcp.md) | MCP server | Hosted self-curating shared memory that keeps your agents working like a high-performing team | vendor's own | | [Granoflow](https://www.anchorterminal.com/tools/granoflow-mcp-server.md) | MCP server | Connect MCP-capable AI agents to local Granoflow tasks, reviews, cards, imports, and work memory. | vendor's own | | [HyperMarrow](https://www.anchorterminal.com/tools/qianshi-hypermarrow.md) | MCP server | Local-first long-term memory for AI coding agents. Stored on your own machine. | vendor's own | | [in-memoria](https://www.anchorterminal.com/tools/pi22by7-in-memoria.md) | MCP server | Persistent codebase intelligence that gives AI assistants memory across sessions | vendor's own | | [kage](https://www.anchorterminal.com/tools/kage-core-kage.md) | MCP server | Verified memory for coding agents: claims cited against code, stale withheld, savings receipts. | vendor's own | | [Kawa Code](https://www.anchorterminal.com/tools/kawacode-mcp.md) | MCP server | Team-aware memory: intent, decisions, real-time conflicts for AI coding assistants. | vendor's own | | [Koragraph](https://www.anchorterminal.com/tools/koragraph.md) | MCP server | Local multi-repo code graph over MCP. Blast radius, call graphs, and memory anchored to code. | vendor's own | | [Kortexio Memory](https://www.anchorterminal.com/tools/kortexio-memory.md) | MCP server | Hosted ambient memory for AI agents. Retrieve context and save notes via MCP tools with OAuth. | vendor's own | | [Lex](https://www.anchorterminal.com/tools/smartergpt-lex.md) | MCP server | Episodic memory, Frames, Policy Neighborhoods, and policy enforcement for AI agents. | vendor's own | | [Lithtrix — Identity, Memory & Trust for AI Agents](https://www.anchorterminal.com/tools/lithtrix-mcp.md) | MCP server | Identity, memory, trust, and swarm primitives for AI agents. MIRC base always free. | vendor's own | | [Living Memory](https://www.anchorterminal.com/tools/viibe-living-memory.md) | MCP server | An MCP memory server. One memory your agents share — across models, devices and apps. | vendor's own | | [Lodekeep](https://www.anchorterminal.com/tools/lodekeep-memory.md) | MCP server | Memory MCP server for Claude Code. Correct a fact and the old version stops being recalled. | vendor's own | | [logic-server](https://www.anchorterminal.com/tools/nocturnus-logic-server.md) | MCP server | Agent reasoning, memory, and token-optimized context for AI applications. | vendor's own | | [Magnemo](https://www.anchorterminal.com/tools/magnemo.md) | MCP server | Governed memory for AI agents. Memory with receipts. Local, plain files, zero network calls. | vendor's own | | [Marrow](https://www.anchorterminal.com/tools/marrow.md) | MCP server | Evidence-grounded, graph-connected, correctable memory for agents. | vendor's own | | [MemCell](https://www.anchorterminal.com/tools/memcell.md) | MCP server | Living memory for AI coding agents: recall before acting, report outcomes so confidence is earned. | vendor's own | | [memento](https://www.anchorterminal.com/tools/runmemento-memento.md) | MCP server | Local-first, LLM-agnostic memory layer for AI assistants. | vendor's own | | [memgit](https://www.anchorterminal.com/tools/memgit.md) | MCP server | Git for AI memory — version-controlled, searchable context that persists across sessions | vendor's own | | [MemoFS MCP Server](https://www.anchorterminal.com/tools/memofs-mcp-server.md) | MCP server | Persistent memory and virtual filesystem MCP server for AI agents. | vendor's own | | [memory](https://www.anchorterminal.com/tools/cortex-mem-memory.md) | MCP server | Shared, persistent memory for AI agents — works with Claude, Kiro, Cursor, or any MCP client. | vendor's own | | [memory-mcp](https://www.anchorterminal.com/tools/letta-memory-mcp.md) | MCP server | MCP server for AI memory management using Letta - Standard MCP format | vendor's own | | [MemoryLayer](https://www.anchorterminal.com/tools/memorylayer.md) | MCP server | Local-first persistent memory for AI agents, with reasoning and formal verification. | vendor's own | | [memxus](https://www.anchorterminal.com/tools/memxus.md) | MCP server | Persistent memory layer that saves and recalls your project context and preferences. | vendor's own | | [mnemodb](https://www.anchorterminal.com/tools/mnemodb.md) | MCP server | Persistent agent memory as plain Markdown in your repo - readable, diffable, with provenance. | vendor's own | | [Neonia: The Cloud Backend for AI Agents](https://www.anchorterminal.com/tools/neonia-cloud-backend.md) | MCP server | Multiple MCP tools, persistent graph memory, token-saving data pointers, and more. | vendor's own | | [Operations Pulse](https://www.anchorterminal.com/tools/openlyuseful-operations-pulse.md) | MCP server | Local-first operational pulses, durable ticket memory, and governed connection discovery. | vendor's own | | [Pastepile](https://www.anchorterminal.com/tools/pastepile-mcp.md) | MCP server | Durable, versioned memory for AI assistants. Remote memory needs a Pastepile subscription. | vendor's own | | [RCLL](https://www.anchorterminal.com/tools/rcll-fleet-memory.md) | MCP server | Self-hosted shared memory for a team of AI agents. Rooms, L0-L3 depth, no LLM on the read path. | vendor's own | | [ReasonGraph Cloud memory](https://www.anchorterminal.com/tools/primaxiom-reasongraph.md) | MCP server | Graph memory for AI agents: entities, cause-effect links, cross-session recall, time travel. | vendor's own | | [recall](https://www.anchorterminal.com/tools/clauderecall-recall.md) | MCP server | Persistent memory for Claude Code and Codex. Search and re-inject context from any session. | vendor's own | | [Reverie](https://www.anchorterminal.com/tools/knowall-reverie.md) | MCP server | Graph memory that dreams: Neo4j knowledge-graph memory for AI agents with hybrid search | vendor's own | | [sdk](https://www.anchorterminal.com/tools/mnemopay-sdk.md) | MCP server | Memory + wallet for AI agents. Real payment rails, Agent FICO 300-850, Merkle audit, identity. | vendor's own | | [stdio-proxy](https://www.anchorterminal.com/tools/undisk-stdio-proxy.md) | MCP server | Undisk MCP: safe file memory for AI agents. | vendor's own | | [Synap Memory](https://www.anchorterminal.com/tools/maximem-synap.md) | MCP server | Persistent memory for AI agents — log and recall conversation context over MCP. | vendor's own | | [Synchronex](https://www.anchorterminal.com/tools/synchronex-mcp-proxy.md) | MCP server | Connect any AI agent to your Synchronex workspace — portable memory, decisions, and workers. | vendor's own | | [Trail](https://www.anchorterminal.com/tools/usetrail-trail.md) | MCP server | Issue tracker and durable project memory for your coding agent, scoped to the folder you open. | vendor's own | | [Tugra](https://www.anchorterminal.com/tools/tugra-ai-tugra.md) | MCP server | Provenance-aware memory format for AI agents: every claim carries its source, age and boundary. | vendor's own | | [wontopos.com MCP server](https://www.anchorterminal.com/tools/wontopos-mcp.md) | MCP server | Wontopos (WOS) long-term memory (beta) - one memory across Claude Code, Cursor, and your agents. | vendor's own | The first 60 of 63, most used first; all of them are in https://www.anchorterminal.com/categories/agent-memory.json and https://www.anchorterminal.com/api/v1/indexed.json (`category`).