Reviva

Local-first AI learning workspace — ask, note, review and create around your own materials. Wiki KB, Agents, Skills, creation tools.Private NotebookLM Altern

What is Reviva?

Local-first AI learning workspace — ask, note, review and create around your own materials. Wiki KB, Agents, Skills, creation tools.Private NotebookLM Alternative.AI 学习工作台,围绕你的资料完成问答、笔记、复习和创作输出。本地优先,多模型,Wiki 知识库,AI Agent,创作工具。本地 NotebookLM

Overview of Reviva

What it is

Reviva is a local-first AI learning workspace designed for studying, research, and knowledge work. It integrates conversation, documents, knowledge bases, notes, review tools, and creation tools into a single desktop application. The core concept is an agent-driven workflow where an AI agent understands materials, calls tools, uses skills, and produces outputs, all while keeping data local. It is positioned as a private alternative to NotebookLM.

Key features

  • Multi-Agent Collaboration: Switch between agents in a conversation; agents can share context and use global memory for long-term preferences.
  • Conversation Around Materials: Select local documents, folders, or Wiki knowledge bases as context; use @ shortcuts and / slash commands to invoke content and skills.
  • Audio/Video Learning: Parse audio and video files for speech transcription, keyframe extraction, and content indexing; ask questions about specific media content.
  • Visual Learning & Creation: 13 creation tools to transform materials into quizzes, flashcards, mind maps, knowledge graphs, charts, PPTs, study guides, and more.
  • Voice Input & Multimodal Models: Voice input support; separate configuration for conversation, speech (ASR), image generation/editing, and OCR models.
  • Connect Existing Knowledge Bases: Built-in KBP protocol to integrate Dify, FastGPT, MCP, or custom HTTP knowledge bases; API keys stored locally.
  • Knowledge Persistence: Upload documents, automatically maintain them as a Wiki knowledge base, and let conversation results flow back into notes and documents.

Use cases

  • Students and exam takers: Organize course materials, handouts, and mistakes; generate quizzes, flashcards, mind maps, and review content.
  • Knowledge workers: Analyze, edit, report, and consolidate knowledge around project materials, meeting notes, and Office documents.
  • Personal knowledge management users: Manage long-term materials, notes, memory, and Wiki with different agents to build a sustainable second brain.

Getting started

Download the latest installer for Windows or macOS from the GitHub releases page or the alternative Quark cloud drive. Choose the version matching your system and processor architecture. Back up important data before upgrading.

License and pricing

The license is not stated in the provided metadata or README.

這是什麼

Reviva 是一個本地優先的 AI 學習工作台,專為學習、研究和知識工作而設計。它將對話、文件、知識庫、筆記、複習工具和創作工具整合到一個桌面應用程式中。核心概念是以 Agent 驅動的工作流程:AI Agent 負責理解資料、呼叫工具、使用技能並產出結果,同時所有資料都保留在本機。它被定位為 NotebookLM 的私有替代方案。

主要功能

  • 多 Agent 協作:在對話中隨時切換 Agent;不同 Agent 可以共享上下文,並使用全域 Agent 記憶保留長期偏好。
  • 圍繞資料對話:直接選擇本機文件、資料夾或 Wiki 知識庫作為上下文;透過 @ 快捷輸入和 / 斜線命令快速呼叫內容與技能。
  • 音視頻學習:解析音訊和視訊檔案,進行語音轉錄、關鍵幀提取和內容索引;在對話中選取媒體後即可針對特定內容提問。
  • 可視化學習與創作:13 個創作工具將資料轉換為測驗、閃卡、心智圖、知識圖譜、圖表、PPT、學習指南等。
  • 語音輸入與多模態模型:支援語音輸入;對話、語音(ASR)、圖像生成編輯和 OCR 四類模型可分別配置。
  • 接入你已有的知識庫:內建 KBP 協定,可整合 Dify、FastGPT、MCP 或自訂 HTTP 知識庫;API Key 儲存在本機。
  • 知識持續沉澱:上傳文件資料,自動維護為 Wiki 知識庫,並讓對話結果回流到筆記和文件中。

使用情境

  • 學生與備考者:整理課程資料、講義和錯題,生成測驗、閃卡、心智圖與複習內容。
  • 知識工作者:圍繞專案資料、會議記錄和 Office 文件完成分析、編輯、報告與知識沉澱。
  • 個人知識管理用戶:用不同 Agent 管理長期資料、筆記、記憶和 Wiki,建構可持續更新的第二個大腦。

快速開始

從 GitHub 發布頁面或備用夸克網盤下載最新的 Windows 或 macOS 安裝包。根據你的系統和處理器架構選擇對應版本。升級前請先備份重要資料。

授權與價格

提供的資料中未說明授權條款。

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/mingchen666reviva.svg)](https://yunodb.com/tools/mingchen666reviva?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/mingchen666reviva?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/mingchen666reviva.svg" alt="Featured on YunoDB" height="20"></a>