Compare
Remnic vs Supermemory.
Supermemory built a good consumer product. Remnic is aimed at a different life: the person running three coding agents who wants all of them to remember the same things, locally.
What Supermemory is
Supermemory is a cloud memory service focused on consumer use: a browser extension, an iOS app, and a save-anything experience for end users. It is polished at what it targets, and it does not target the coding-agent space.
Where Supermemory wins
- Consumer polish: save from the browser, recall on the phone, no daemon anywhere.
- Zero setup for non-developers.
Where Remnic differs
- Built for the agents you code with. Native plugins for Claude Code, Codex CLI, Pi, Hermes, and OpenClaw, plus MCP for Cursor and the rest. Supermemory does not aim there.
- Local files, not a cloud account. Markdown on your disk, MIT, free forever. Your memory does not live in someone else's product decisions.
- Credit where due. Remnic borrows one idea from Supermemory's ASMR work: parallel specialized retrieval, where DirectFact, Contextual, and Temporal retrievers run concurrently and total latency equals the slowest rather than the sum. Good ideas get adopted and attributed.
- Correction and provenance. Recall X-ray explains every result; MemCorrect measures whether corrections stick. See benchmarks.
Switching
Export your data from Supermemory as JSON, then import it with provenance metadata and a dry-run preview:
remnic import --adapter supermemoryFull walkthrough on the migration page and in the Supermemory import guide.