We’re building ultra-lightweight, embeddable components for modern AI infrastructure—starting with vector search, and expanding into RAG, memory systems, and edge-native AI workflows.
Think of us as your friendly neighborhood team making it easy to bring semantic intelligence directly to apps, devices, and agents—without servers, without the cloud, and without compromise on privacy or speed. 🚀
| Project | What it does |
|---|---|
| zvec | A lightweight, lightning-fast, in-process vector database. |
| zvec-web | The official web frontend and documentation portal for Zvec. |
| zvec-node | High-performance Node.js bindings for Zvec. |
| zvec-mcp-server | A Model Context Protocol (MCP) server for Zvec. |
| zvec-agent-skills | Official Agent skills for building with Zvec. |
More tools are on the way—stay tuned! 🛠️
We’re open-source first, and we thrive on real-world feedback:
- ✨ Star zvec if you like what you see
- 🐞 Found a bug? Open an issue!
- 💡 Have a use case? Share it—we’re collecting real RAG/memory scenarios (and yes, there are swag rewards!)
- 🤝 Want to add a binding (Rust? Go?)? Let’s chat in Discussions!
Your input shapes where we go next—whether it’s better quantization, DuckDB integration, or on-device memory agents.
Stay updated and get support — scan or click:
| 💬 DingTalk | 🎮 Discord | X (Twitter) | |
|---|---|---|---|
 |
 |
 |
 |
| Scan to join | Scan to join | Click to join | Click to follow |