MemOS Model Card
这是可索引项目报告证据页：它保留 MemOS Model Card 的源材料入口、机制线索和限制提醒；正文仍需 reader/editor 与 academic public-copy review 后才能当作最终结论引用。
MemOS Model Card
| Field | Value |
|---|---|
| Repository | memtensor/memos |
| Category | Self-Evolving Memory OS |
| Stars / forks snapshot | 9400 / 846 |
| Language | TypeScript/Python |
| License | Apache-2.0 |
| Raw capture | raw-github/memtensor_memos.md |
| Updated by | hourly public metadata update, 2026-05-25 |
1. Role in Self Evolve
MemOS 是面向 LLM 和 Agent 的 memory operating system，把 memory API、memory cube、插件、dashboard 和 self-evolving memory layers 合成可治理的长期记忆基础设施。
2. Working Principle
agent event/tool trace -> memory cube/API -> hybrid retrieval/governance -> skill/world-model crystallization -> reusable long-term memory
3. Evidence Path
web GitHub page observed 1,778 commits, Apache-2.0 license, TypeScript/Python stack, release v2.0.16 latest May 19 2026, 9.4k stars and 846 forks; README reports +43.70% accuracy vs OpenAI Memory, 35.24% token savings, LoCoMo/LongMemEval/PrefEval/PersonaMem signals, and self-evolving memory L1-L3 plus Skills. Shell GitHub API access remained blocked by DNS and local gh auth was invalid, so this card treats the current snapshot as web-observed rather than API-verified.
4. Teaching Use
Use this card to explain Self-Evolving Memory OS in the raw -> classification -> project card -> site/report pipeline. The reading path is: raw capture -> classification row -> public site card -> project report -> aggregate GitHub analysis.
5. Limits
当前未克隆源码，未运行 benchmark、SDK examples、skill install flows、memory experiments、agent evolution loops 或 production deployments；star/fork/commit 快照来自公开 GitHub 页面文本或可见页面片段。