AI工具 Skill技能库 关于道场
Swift AI工具 开源 ai-agents cli coreml coreml-framework data-science machine-learning mcp mcp-server memory memory-cache memory-hacking metal on-device-ai rag rag-pipeline vector-database vector-embeddings vector-search vectordb

Wax:文件内存检索增强生成在本地运行服务器

GitHub |2026-07-02

Single-文件 内存 layer 为 AI agents, sub mili-second 检索增强生成在 Apple Silicon. Metal Optimized 本地运行. No 服务器. No API. One 文件. Pure SwiftGitHub 767 ⭐ / 47 Fork,主要使用 Swift 开发。标签:ai-agents,cli,coreml,coreml-framework,data-science。访问项目:https://github.com/christopherkarani/Wax

项目介绍

Single-文件 内存 layer 为 AI agents, sub mili-second 检索增强生成在 Apple Silicon. Metal Optimized 本地运行. No 服务器. No API. One 文件. Pure Swift

核心功能

- Persistent chatbots — Your assistant remembers every conversation, preference, and decision across sessions. - Coding agents with long-term memory — Claude Code or Cursor that recalls your codebase patterns, architectural decisions, and TODOs from last week. - Personal knowledge bases — Semantic search over your notes, documents, and web clips. Ask "What did I read about HNSW?" and get the exact paragraph. - On-device RAG — Ship AI features in your iOS or macOS app without calling the cloud.

使用指南


项目信息

项目信息
项目名称christopherkarani/Wax
Star 数767
Fork 数47
编程语言Swift
开源协议Apache-2.0
创建时间2026-01-20
最近更新2026-06-29
主题标签ai-agents、cli、coreml、coreml-framework、data-science

项目地址

GitHub 仓库:<https://github.com/christopherkarani/Wax>

> 完整 URL:[https://github.com/christopherkarani/Wax](https://github.com/christopherkarani/Wax)

如需查看完整源码、安装命令、使用示例、API 文档,请直接访问上方项目地址。所有信息均以 GitHub 仓库为准。