展示HN:一款适用于macOS的本地离线文档聊天应用程序

1作者: navid72m大约 1 个月前原帖
嗨,HN, 我开发了一款 macOS 桌面应用,让你可以完全在本地与文档进行聊天。 没有云服务,没有 API 密钥,没有数据离开你的机器。所有功能都在 Apple Silicon 上离线运行,使用 GGUF 模型和 llama.cpp。 它的功能包括: - 上传 PDF、文本文件和图像 - 对图像和扫描的 PDF 进行光学字符识别(OCR) - 本地嵌入和检索(RAG) - 使用本地大语言模型(LLM)与文档聊天 - 模型在首次运行时下载并存储在本地 技术栈: - Electron(前端) - Python 后端打包为本地二进制文件 - llama.cpp + GGUF(目前使用 Gemma / Mistral 类模型) - SentenceTransformers 用于嵌入 - FAISS 用于向量搜索 - 完全在设备上运行(CPU / Metal) 我为什么要开发这个: 我想要一个以隐私为优先的替代方案来替代云文档聊天工具。将完整的本地 LLM + OCR + RAG 流水线打包成一个单一的 macOS 应用,结果比预期要困难得多(Gatekeeper、PyInstaller、动态库、模型大小等)。 下载链接: GitHub 发布(macOS Apple Silicon): [https://github.com/navid72m/chatbot/releases/tag/v.0.1.2](https://github.com/navid72m/chatbot/releases/tag/v.0.1.2) 关于 macOS 安全性的说明: 由于该应用尚未签名,macOS 可能会在首次启动时阻止它。 你可以运行以下命令: `xattr -rd com.apple.quarantine "/Applications/Document Chat.app"` 我非常希望能收到关于以下方面的反馈: - 文档聊天的用户体验(UX) - 模型选择 / 性能 - 其他人如何在桌面上处理本地 RAG 欢迎提出技术问题。
查看原文
Hi HN,<p>I built a macOS desktop app that lets you chat with your documents completely locally.<p>No cloud, no API keys, no data leaving your machine. Everything runs offline on Apple Silicon using GGUF models and llama.cpp.<p>What it does: - Upload PDFs, text files, and images - OCR for images and scanned PDFs - Local embeddings + retrieval (RAG) - Chat with documents using a local LLM - Models are downloaded on first run and stored locally<p>Tech stack: - Electron (frontend) - Python backend bundled as a native binary - llama.cpp + GGUF (currently Gemma &#x2F; Mistral class models) - SentenceTransformers for embeddings - FAISS for vector search - Runs entirely on-device (CPU &#x2F; Metal)<p>Why I built this: I wanted a privacy-first alternative to cloud document chat tools. Packaging a full local LLM + OCR + RAG pipeline into a single macOS app turned out to be much harder than expected (Gatekeeper, PyInstaller, dylibs, model size, etc.).<p>Download: GitHub release (macOS Apple Silicon): <a href="https:&#x2F;&#x2F;github.com&#x2F;navid72m&#x2F;chatbot&#x2F;releases&#x2F;tag&#x2F;v.0.1.2" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;navid72m&#x2F;chatbot&#x2F;releases&#x2F;tag&#x2F;v.0.1.2</a><p>Note on macOS security: Because the app is not signed yet, macOS may block it on first launch. You can run: xattr -rd com.apple.quarantine &quot;&#x2F;Applications&#x2F;Document Chat.app&quot;<p>I’d really appreciate feedback on: - UX for document chat - Model choices &#x2F; performance - How others approach local RAG on desktop<p>Happy to answer technical questions.