Head to head
6 of 12 recorded facts differ. Everything below is source-reported with a capture date. Niched has not published a buyer verdict for this pair, so nothing here declares a winner. A winner only exists inside a defined buyer niche with stated constraints.
An open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of tasks that could take. deer-flow is open source, written primarily in Python, released under MIT License. Its repository has 78.2K GitHub stars and 10.7K forks, and was last committed to within the last two weeks.
Visit ↗Self-evolving memory OS for LLM & AI Agents: ultra-persistent memory, hybrid-retrieval, and cross-task skill reuse, with 35.24% token savings
Visit ↗| Fact | deer-flow | MemOS |
|---|---|---|
| Official SDK installs, 30 days | Not recorded | Not recorded |
| SDK evidence captured | Not recorded | Not recorded |
| GitHub stars | 78.7K | 10.5K |
| Forks | 10.7K | 966 |
| Open issues | 938 | 90 |
| Last commit | yesterday | 2 days ago |
| Primary language | Python | TypeScript |
| License | MIT License | Apache License 2.0 |
| Open source | Yes | Yes |
| Pricing model | Not recorded | Not recorded |
| Founded | Not recorded | Not recorded |
| Categories | AI agents & autonomous workflows | AI agents & autonomous workflows |
Both products are placed in AI agents & autonomous workflows. A category organises the market; it carries no score. See all deer-flow alternatives.