Open-source Fabric alternatives

AppFlowy is an open-source workspace that brings documents, wikis, project and task management and team collaboration into one app. It is presented as an alternative to Notion, and its topics also tag it as a Confluence alternative, with the emphasis on keeping control of your own data.

The client is built with Flutter and a Rust core, giving apps for macOS, Windows and Linux plus iOS and Android versions. Desktop builds can be downloaded directly or from channels such as Flathub and Snapcraft. AI features are part of the workspace, and a public roadmap and active forum let users propose features. Translations are contributed through JSON files and tools such as inlang.

The project is AGPL-3.0 licensed and has documentation for self-hosting the server side, so organizations can keep content on their own infrastructure. A hosted version is available through appflowy.com for people who do not want to run servers. It suits teams looking for a Notion-style workspace with a privacy-first stance.

AFFiNE is an open-source workspace that combines writing, drawing and planning in a single app. Documents, an edgeless whiteboard and table views share the same blocks, so you can place rich text, sticky notes, embedded web pages, database views, linked pages, shapes and even slides anywhere on one canvas. It is presented as an alternative to Notion and Miro.

The project stresses local-first storage, meaning your data lives on your own device first, together with real-time collaboration. A multimodal AI assistant helps write reports, turn outlines into slides, summarize articles into mind maps and organize tasks. Under the hood it uses CRDTs for syncing and is built mostly in TypeScript and Rust, with Electron for the desktop app. It also supports Markdown, wiki and knowledge-base use, and presentations.

AFFiNE's license is listed as 'Other' on GitHub, so review it before commercial use. There is a live demo and documentation on the project site, a hosted version is available, and the app can also be self-hosted.

SiYuan is an open-source, privacy-first personal knowledge management system. Its distinctive feature is fine-grained, block-level referencing: any block can be linked to or embedded elsewhere, with two-way links, custom attributes and SQL query embeds. Editing is Markdown WYSIWYG, and the project describes itself as a workspace where humans and AI agents work together. It is local-first and self-hostable.

The editor supports block-style writing, list outlines, block zoom-in, very large documents, math formulas, charts, flowcharts and Gantt charts. Web clipping, PDF annotation links, flashcards with spaced repetition, a database table view, Tesseract OCR and multi-tab split-screen layouts are included. You can export blocks and embeds, standard Markdown with assets, or PDF, Word and HTML. AI writing and Q and A chat can use the OpenAI API, and template and JavaScript or CSS snippets allow customization.

SiYuan is AGPL-3.0 licensed and available as desktop apps, mobile apps for Android, iOS and HarmonyOS, an installation package, a package manager, and Docker, Kubernetes, Unraid and TrueNAS hosting. A command line interface and an app market for extensions exist. Most features are free, even for commercial use.

Open Notebook is a privacy-oriented, open-source option for people who want something like Google's NotebookLM for researching and learning with AI. It is written in TypeScript and released under the MIT license. The project describes itself as private, multi-model and fully local, so your research material stays on infrastructure you control instead of a single cloud provider.

It supports more than a dozen AI providers, including OpenAI, Anthropic, Ollama and LM Studio, and lets you organize multimodal content such as PDFs, videos, audio and web pages. Features include podcasts with several speakers, keyword and vector search over everything you add, and chat conversations grounded in your own sources. The interface is available in several languages, including Portuguese, Chinese, Japanese, Russian and Bengali.

The README compares Open Notebook with Google NotebookLM across privacy and control, choice of AI provider and more, and links a getting-started guide, user guide, feature list and deployment instructions. A Discord server is open for help and workflow ideas. It suits students, researchers and knowledge workers who want a NotebookLM-style workflow with their own choice of models.

NoteGen is a local-first Markdown note-taking app built around the idea of capturing first and organizing later. It helps collect scattered snippets, records and clippings and then uses AI to arrange them into clear notes, acting as a personal knowledge base.

It is a desktop application built with Tauri and Next.js. The project topics point to Markdown editing, WebDAV sync, retrieval augmented generation, MCP and agent features, and the README describes configurable model settings with an OpenAI-compatible interface for conversations, note organization, writing assistance, translation and summarization. Knowledge base features reference embeddings, reranking, OCR and speech.

NoteGen is licensed under GPL-3.0, and downloads, documentation, a business offering and community discussions are linked from the repository. Since it is a local app, you keep notes on your own machine and choose which model provider to connect, which suits people building a personal knowledge management system.

Anytype Desktop is the official desktop client for Anytype, a personal knowledge base for macOS, Windows and Linux. It lets you collect pages, tasks, wikis and journals and link them together, while keeping data stored locally and encrypted across your devices.

The client is built around composable blocks such as text, databases, kanban boards and calendars, and you can define your own Types to shape the data model. Storage is offline-first with optional peer-to-peer sync, and encryption relies on the any-sync protocol. It is an Electron application written in TypeScript, and it can be extended through a gRPC API and AI agents described in the repository.

Sign-in uses an Any-ID, and installers are published on the releases page and the download site. The code is published under the Any Source Available License 1.0 rather than a standard open-source licence, and the README includes instructions for building from source and updating the middleware.

Blinko is a card-style note-taking app with AI features, designed for people who want to capture fleeting ideas the moment they happen. It is self-hosted and puts privacy first, so your notes stay on a server you control. The interface is built with Next.js and React in TypeScript, backed by PostgreSQL.

Its AI-enhanced retrieval uses retrieval-augmented generation, so you can find notes by asking in natural language rather than remembering keywords. Notes use Markdown and the project is tagged for memos and notebooks. A live demo, official site, documentation and English and Chinese Telegram communities are linked from the README.

Blinko is licensed under GPL-3.0 and installed on your own server, typically with Docker. It suits individuals who want a private alternative to cloud note and memo apps and are comfortable connecting their own AI provider.

EdgeEver is an open-source notes and knowledge base workspace that revives the three-pane layout familiar from Evernote while adding an open data architecture and integration with AI agents. It is presented as a portable Evernote alternative, is written in TypeScript, and is released under the AGPL-3.0 license.

It can run serverlessly within Cloudflare's free quotas using Workers, D1, and R2, so no server purchase is needed, or you can deploy the same application with Docker on a VPS, NAS, or home server. Native MCP support lets AI agents work with your notes. Other topics include a web clipper, offline-first behavior, PWA and Electron clients, SQLite storage, and the Tiptap editor.

A live demo and client downloads are provided, with the interface available in several languages. EdgeEver was created in 2026, so it is young. It suits people migrating from Evernote and developers who want a knowledge base that agents can read and write.

Eclaire is an open-source, privacy-focused AI assistant that works across your personal data: tasks, notes, documents, photos, bookmarks and more. It aims to be a complete product rather than a framework, assembling proven building blocks into something you can run on your own hardware with local models and local data storage. It targets people who want AI help with personal or sensitive information without using hosted services.

The README lists four ideas: self-hosting, unified data so the AI can see and connect everything in one place, AI-powered features such as content understanding, search, classification, OCR and automation, and open-source extensibility through APIs. It also has sections on selecting models, architecture and development.

The project carries two prominent warnings. It is pre-release and under active development, so breaking changes are expected and backups are advised. It must also not be exposed directly to the public internet, and the authors recommend placing it behind Tailscale, Cloudflare Tunnels or an authenticating reverse proxy. Eclaire is written in TypeScript and licensed under MIT.