YourMemory icon

YourMemory

YourMemory is a persistent memory system for AI agents that helps assistants retain useful context across sessions. It runs locally, supports MCP compatibility, and includes tools for memory review, audit logging, and cost analysis.

YourMemory

Persistent memory for AI agents

YourMemory is a persistent memory system for AI agents. It is designed to help assistants retain useful information across sessions instead of resetting to zero after every conversation.

The product describes its memory as biologically inspired: memories strengthen when recalled, fade when unused, and connect to one another through a hybrid graph and vector retrieval approach. The site positions this as a way to keep long-term context available without carrying forward everything indefinitely.

The homepage emphasizes practical workflow support. Users can estimate token savings with a prompt-based audit, review stored memories in a browser, and inspect a tamper-evident audit trail of memory operations. A free activation flow is available, and the site says the system runs locally with no API key required.

For teams, YourMemory also offers an Enterprise build for shared memory across an organization. The site frames this as a way to preserve solved issues, customer details, and architectural decisions so later sessions can surface relevant knowledge automatically.

Core features

Biologically inspired memory decay

YourMemory models memory as something that can strengthen when recalled and fade when left unused, instead of storing every item indefinitely. The homepage presents this as a lifecycle that keeps only the most relevant information active.

Hybrid graph and vector retrieval

The product combines graph-style linking with vector retrieval so memories can connect across sessions rather than remain isolated chunks. The site contrasts this with legacy vector databases that return chunks without prioritization or forgetting.

Measured benchmark reporting

The homepage highlights benchmarked results across three public datasets and calls out 89.4% Recall@5 on LongMemEval in the meta description. It also says the results are reproducible and that benchmark code is available in the repo.

Memory cost and savings audit

YourMemory includes an instant audit prompt that can be pasted into Claude, ChatGPT, Gemini, Cursor, or Copilot to estimate savings from a conversation. It can also calculate token and cost impact from an exported chat history in the browser.

Memory browser and audit trail

The interface includes a Memory Browser for reviewing stored memories, strength, category, and recall history, plus an Audit tab that logs reads, writes, and deletes in a tamper-evident hash chain.

Local activation and setup

The site offers a free activation flow that runs locally, requires no API key, and provides a token for command-line setup. It also links to a GitHub repo and Discord community.

Common use cases

  • Long-running assistant sessions

    Use YourMemory when you want an assistant that remembers prior preferences, decisions, or project details after the current chat ends. The product is built around carrying relevant context forward across sessions.

  • Reviewing stored memory and recall history

    Use the Memory Browser and audit trail when you need to review what was stored, how strong a memory is, and when it was last recalled. This is useful for checking why a memory appears in later sessions.

  • Token and cost analysis

    Use the cost audit flow before or after a conversation to estimate how much context you are re-sending over time. The site offers both a quick prompt-based check and a browser-side calculation from exported chat history.

  • Shared team knowledge

    Use the Enterprise build when a team wants shared memory for repeated customer issues, architecture decisions, or internal knowledge. The homepage shows shared and private memory states and describes the enterprise version as a private build.

  • Local agent setup

    Use the setup flow if you want a local installation that does not depend on an external API key. The site says activation takes under two minutes and provides a command-line token after email verification.

Pros and Cons

Pros

  • Designed for persistent memory across sessions rather than one-off chat context.
  • Uses a memory lifecycle that includes recall and natural fading, which may reduce stale context.
  • Provides visible tooling for memory inspection, recall history, and audit logging.
  • Supports a local activation flow with no API key required.
  • Includes an Enterprise option for shared team memory.

Cons

  • The pricing page is not available on the site, so the pricing structure is unclear.
  • Integration details beyond MCP compatibility and a few example apps are limited.
  • The site gives benchmark claims, but the homepage does not provide a full technical breakdown of model support or deployment requirements.

FAQ

How do I start using YourMemory?

YourMemory offers a free activation flow and runs locally on your machine. The site says it starts in under 2 minutes and does not require an API key.

What tools does it work with?

The homepage describes YourMemory as compatible with MCP and shows example use with Claude, ChatGPT, Gemini, Cursor, and Copilot in its memory-audit workflow. No broader integration catalog is listed on the site.

How does the memory system work?

The product is built to learn from a session, store linked memories, recall relevant items later, and let memories fade naturally when they are no longer used. The site presents this as a three-step lifecycle: memory is born, stored and linked, recalled, and then fades.

Can teams use YourMemory?

Yes. The site includes an Enterprise version for team memory sharing, with shared and private memory shown in the interface examples. It is described as a private build and requires contacting the team to get started.

Is there evidence for the performance claims?

The homepage cites benchmark results on public datasets and mentions 89.4% Recall@5 on LongMemEval in the site description. The site also says the benchmark code lives in the repo, but no separate technical paper is included on the page.

Quick Facts

Category
Developer Tool
Primary use
Persistent memory for AI agents
Compatibility
MCP compatible
Deployment
Runs locally on the user's machine
Activation
Free activation flow
Source domain
yourmemoryai.xyz

YourMemory Alternativen

Falconer icon

Falconer

Falconer is an AI-powered company brain that keeps internal documentation accurate, searchable, and up to date by syncing with tools like GitHub, Slack, and Linear. It helps engineering and cross-functional teams centralize knowledge and answer questions from a shared source of truth.

Goldfish icon

Goldfish

Goldfish is an AI memory app for macOS that helps you reply, write, summarize, and continue using the context already on your Mac. It stores memory locally and writes in your tone.

Lasso icon

Lasso

Lasso is an ecommerce product data platform for enriching catalog records, processing supplier files, generating product content, and monitoring competitors. It combines a web app with a REST API, SDK, and MCP server for teams and developers.

CreateOS Sandbox icon

CreateOS Sandbox

CreateOS Sandbox is an isolated compute environment for running code and agent workloads inside Firecracker micro-VMs. It is designed for workflows that need machine-level isolation, private networking between sandboxes, and programmatic control through SDK, CLI, or MCP.

Codex Plugins icon

Codex Plugins

Codex Plugins bundle reusable skills, app integrations, and MCP servers into workflows you can install in the Codex app or use from Codex CLI. They help extend Codex with connected-service tasks, reusable instructions, and shared team workflows.

Struere icon

Struere

Struere is an AI-native platform for turning spreadsheet data into structured operational software with dashboards, alerts, and automations. It is aimed at teams that want to replace manual spreadsheet workflows without building custom tools from scratch.