u14app/neo-chat

A local-first AI chat workspace for models, agents, skills, plugins, search, RAG, voice, memory, and artifacts.

What it solves

Neo Chat provides a self-hostable, local-first AI workspace that allows users to maintain ownership of their data. It eliminates the need to rely on centralized AI platforms by keeping chat history, memories, and knowledge bases stored in the browser, while using server routes as controlled proxies for AI model providers and external tools.

How it works

The application is built with Next.js, React, and TypeScript. It uses a local-first architecture where metadata and messages are stored in IndexedDB and files are kept in the browser's Origin Private File System (OPFS). For synchronization across devices, it offers an optional end-to-end encrypted vault using WebDAV or S3/MinIO. It connects to various AI providers (Google, Anthropic, OpenAI) and supports the Model Context Protocol (MCP) for extending functionality via remote servers or a local Docker bridge.

Who it’s for

It is designed for users who want a powerful, feature-rich AI chat interface with advanced capabilities like RAG and agentic workflows, but who prioritize data privacy and local ownership of their AI interactions.

Highlights

  • Local-First Storage: Chat history, memories, and search indexes stay in the browser by default.
  • Advanced AI Capabilities: Supports multi-provider chat, agent mode for tool-calling models, and knowledge-base RAG with hybrid lexical/vector retrieval.
  • Extensible Tooling: Integrates OpenAPI-style plugins, remote MCP servers, and parameterized "Skills".
  • Privacy-Focused Sync: Opt-in end-to-end encrypted synchronization across devices.
  • Rich Media Support: Native image generation/editing, multimodal image input, and voice input/output.
  • Deployment Flexibility: Can be deployed via Docker, Vercel, or Cloudflare Workers.

Related

  • Project
    Lucassssss/eechat

    A secure local AI chat application that supports the Model Context Protocol (MCP) for easy integration of AI tools and services while keeping data stored locally.

  • Project
    hAcKlyc/MyAgents

    An open-source desktop workspace for personal AI agents that integrates local files, terminals, and a task management system to move beyond simple chat interfaces.

  • Project
    senbo1/chat0

    An open-source, privacy-focused AI chat application that lets users bring their own API keys to interact with multiple AI models via a fast React interface.

  • Project
    ThinkInAIXYZ/deepchat

    DeepChat is an open‑source, cross‑platform desktop client that lets you run LLM‑based agents locally. It unifies cloud and Ollama models, records every interaction with a recoverable “Tape” system, supports reusable Skills, ACP‑compatible agents, and remote control via Telegram, Discord, Feishu, QQBot and WeChat. The app offers rich markdown/code rendering, built‑in web‑search extensions, strong MCP tooling, and privacy‑focused storage, all under an Apache‑2.0 license.

  • Project
    llt22/talkio

    A local-first, multi-model AI chat client that enables group conversations between different AI models with customizable personas and MCP tool integration.