cachly vs claude-mem: two memories for Claude Code
Both give Claude Code a memory across sessions. claude-mem records what the agent does in each session, compresses it with an AI model and keeps it on your machine. cachly keeps lessons from fixes and decisions in a hosted brain in Germany, reachable from every machine you work on and, on a team plan, from your whole team.
This page compares both on facts you can check. Each row links its source.
At a glance
| Aspect | cachly | claude-mem | Sources |
|---|---|---|---|
| What it is | cachlyExperience for coding agents. Your assistant stores a lesson after a fix and finds it again when the same problem returns. | claude-memA memory system built for Claude Code. It captures what the agent does in a session, compresses it with an AI model, and injects relevant context into later sessions. | Sources |
| How memory is written | cachlyA lesson: topic, what worked, what failed, context, files and commands. Your assistant writes it, the Stop hook records reported fixes, and git history adds more. | claude-memAutomatically. Hooks record tool use as observations, and an AI model turns them into summaries. | Sources |
| Which model does the work | cachlyYour assistant writes each lesson in its own session. cachly computes the embeddings for search. | claude-memYour choice: the hosted claude-mem observer (CMEM Pro), your own OpenRouter or Gemini key, or your Anthropic plan. | Sources |
| Where data lives | cachlyHosted by cachly: one instance per brain at Hetzner in Nuremberg, Germany. By default, cachly computes embeddings in the EU. The API runs behind Cloudflare, a US company. | claude-memOn your machine: a SQLite database in ~/.claude-mem and a Chroma vector database. With CMEM Pro, synced observations are also stored by CMEM. | Sources |
| Claude Code hooks | cachlySessionStart, UserPromptSubmit and Stop. | claude-memSessionStart, UserPromptSubmit, PostToolUse, Stop and SessionEnd. | Sources |
| MCP tools | cachlyAn MCP server on your machine. The editor lists 27 tools; 124 in total through one dispatcher tool. | claude-memSearch tools: search, timeline and get_observations. They fetch an index first and full details only for chosen IDs. | Sources |
| Coding tools | cachlyClaude Code as a plugin. Cursor, Windsurf, VS Code with GitHub Copilot, Cline, Continue.dev and Zed over MCP. A JetBrains plugin. | claude-memClaude Code, plus installers for OpenCode, T3 Code (Codex and Claude Code), Antigravity CLI, OpenClaw and others. | Sources |
| On your machine | cachlyThe MCP server and the hooks. Needs Node.js 20 or later. Storage and search run on cachly's servers. | claude-memA worker service managed by Bun, with SQLite and Chroma. Needs Node.js 20 or later; Bun and uv are installed when missing. | Sources |
| Teams | cachlyTeam plan: €49 a month for 5 seats. A lesson one developer stores reaches the whole team. | claude-memTeamBrain is in development with design partners. Public pricing is not announced yet. | Sources |
| License | cachlyMCP server and hooks: Apache-2.0. The hosted service is closed source. | claude-memApache-2.0. | Sources |
| Pricing | cachlyFree €0: one brain, 25 MB, 500 recalls a month. Dev €19, Pro €49, Business €199 a month. Team: €49 a month for 5 seats. | claude-memThe engine is free and open source. CMEM Pro adds hosted observation and cloud sync: $30 a month after a 30-day trial. Heavy $70, CMEM Max $100 a month. | Sources |
| Community | cachlycachly-mcp: public repository since April 2026. | claude-memPublic repository since August 2025, with a large open-source community. | Sources |
As of . Every statement about claude-mem comes from the linked source, read on that day. Products change; if you spot something outdated, write to [email protected].
When to pick claude-mem
- You want memory on your own machine. The open-source engine keeps it locally, in SQLite and Chroma.
- You want memory without writing anything down. Its hooks record every session on their own.
- You want to browse past sessions. A local web viewer shows the memory stream as it grows.
- You value a large, established open-source community.
- You want the whole engine under Apache-2.0.
When to pick cachly
- You want each lesson in a fixed form: the fix first, then what failed on the way, the files and the commands.
- Your team should share lessons on a plan with a public price: €49 a month for 5 seats.
- You switch between editors and want the same lessons in Cursor, Windsurf, VS Code with Copilot or JetBrains.
- Your lessons should live on a server in Germany, reachable from every machine you work on.
- You want to try it before you create an account.
Try cachly
In Claude Code, cachly comes as a plugin with three hooks. Try it without an account: the first tool call creates a test brain on EU servers. The commands are in Hooks that carry experience.
For Cursor, Windsurf, VS Code and other editors, MCP integration shows the one-command setup. What leaves your machine, and where it lands: EU hosting and your data.
Sources
Read in the original text on .
Other comparisons: cachly vs Supermemory · cachly vs Zep · cachly vs Basic Memory · All docs