# Comparisons

> How xmemory compares with other agent memory products and with the architectural patterns teams reach for instead. Short, evidence-linked pages, one per alternative, with an XMD schema showing the equivalent in xmemory.

Canonical page: https://xmemory.ai/compare/

Each page below follows the same shape: what the alternative does, where xmemory is stronger and
where it is not the point, the evidence we have and its limits, an XMD schema that expresses the
alternative's memory model in xmemory, and when to choose which. Documentation for every
alternative was reviewed on the date stated at the top of its page.

## Agent memory products

- [xmemory vs Mem0](https://xmemory.ai/compare/mem0/index.html.md): schema-defined records, key-based updates and SQL queries
  versus extracted memory text and ranked retrieval; includes a Mem0-style XMD schema.
- [xmemory vs Zep](https://xmemory.ai/compare/zep/index.html.md): schema-resolved records versus a temporal knowledge graph and
  context block; what each type system constrains and how corrections land.
- [xmemory vs Cognee](https://xmemory.ai/compare/cognee/index.html.md): an instance-owned schema resolved on every write versus
  a document-to-graph pipeline with per-call graph models.
- [xmemory vs Letta](https://xmemory.ai/compare/letta/index.html.md): schema-owned records versus a git-backed Markdown memory
  filesystem the agent edits; contract versus convention.
- [xmemory vs Supermemory](https://xmemory.ai/compare/supermemory/index.html.md): an executable schema versus context-string
  steering over a temporal fact graph and profiles.
- [xmemory vs Graphlit](https://xmemory.ai/compare/graphlit/index.html.md): a memory of agent state versus a content platform of
  Schema.org observables linked to sources.
- [xmemory vs LangMem and the LangGraph Store](https://xmemory.ai/compare/langmem/index.html.md): resolution rules enforced for
  every writer versus model-driven consolidation over a JSON document store; complementary via the
  LangChain integration.
- [xmemory vs model-vendor context primitives](https://xmemory.ai/compare/model-vendor-memory/index.html.md): one schema-governed
  memory across apps and vendors versus Anthropic's memory tool, OpenAI conversation state and
  Gemini context caching.
- [xmemory vs Hindsight](https://xmemory.ai/compare/hindsight/index.html.md): domain-typed records versus world and experience
  facts consolidated into observations with an agentic reflect step.
- [xmemory vs Honcho](https://xmemory.ai/compare/honcho/index.html.md): schema-defined records and relational queries versus
  reasoning-maintained representations of peers; complementary.

## Architectural patterns

Pages on the "why not just…" alternatives (vector store and RAG, knowledge graphs and GraphRAG,
long context, transcript and summary memory, DIY tables, files as memory, coding-agent harness
memory) are being added here. Until then, the [Why page](https://xmemory.ai/why/) covers text and Markdown memory,
graph memory and MCP over a database.

## Why we think this matters, and for whom

xmemory works well for short-running agents, and it is at its best for repeated tasks, where each run adds to a body of knowledge the next run can use. Long-horizon agents deserve a special mention, because there is very little objective evidence yet about what memory design works for them. Our working belief, and the reason xmemory exists: an agent running a long, multi-step task does better on structured state than on accumulated context. A schema gives it an explicit feedback loop at write time (what was accepted, what was rejected, what changed), so memory is updated rather than bloated and the trajectory stays on course: fewer re-reads, faster steps, more reliable outcomes.

If that is your workload, try xmemory on it. We are building objective measurements for this claim. The [Toloka evaluation](https://xmemory.ai/toloka-agent-memory-evaluation/) is our first independent measurement of the benefits of structured memory; evaluating changing working state over long-horizon tasks is next, and we would welcome collaborators on it. Write to [info@xmemory.ai](mailto:info@xmemory.ai).