LobeHub
LobeHub, formerly LobeChat, is a self-hostable workspace for building and running AI agents, with an agent builder, groups of agents that work together, scheduled runs, editable memory, and a library the README puts at over 10,000 tools and MCP plugins. Its LobeHub Community License is Apache-2.0 with added conditions: commercial use is allowed unmodified, but distributing a modified version needs a paid commercial licence, so this site labels it source available.
What does LobeHub do?
LobeHub, the repository formerly called lobe-chat, describes itself as "a work-and-lifestyle space to find, build, and collaborate with agent teammates that grow with you." The unit of work is the agent: an agent builder creates them, agent groups have them work together, and Schedule runs them at set times "even while you are away." Pages lets several agents write with shared context, Projects and Workspaces organise work for teams, and an IM gateway puts agents in the chat apps you already use.
It lists "10,000+ Skills", a library of tools and MCP-compatible plugins, extending function calling through a plugin index, template, SDK and gateway. Personal memory is "structured, editable memory" you can inspect. The README's setup marks an OpenAI API key as required and says an API key is needed to chat with models; it does not name Ollama or Anthropic. It does not describe document chat, web search, image generation or mobile apps.
Key facts
- Licence: the LobeHub Community License, based on Apache-2.0 with added conditions. Commercial use as a front end and back end without modifying the source is allowed; developing and distributing a derivative work requires a commercial licence from LobeHub.
- Contributors agree that LobeHub may change the licence terms and use contributed code commercially, including in its cloud edition. GitHub reports the licence as Other.
- Agents as the unit of work: an agent builder, agent groups, Pages, Schedule, Projects and Workspaces.
- A library the README counts at over 10,000 tools and MCP-compatible plugins.
- Structured, editable personal memory, and an IM gateway to chat apps.
- The setup example requires an OpenAI API key; other providers are not named in the README.
- Deploy: one-click buttons for Vercel, Zeabur, Sealos, RepoCloud and Alibaba Cloud, or Docker through a setup script and Docker Compose; a hosted app runs at app.lobehub.com.
- The repository moved from lobehub/lobe-chat to lobehub/lobehub. Latest tagged release when read: v2.2.17 on 2026-09-11; 82,567 stars on 2026-09-17.
What does it replace, and where does it fall short?
LobeHub replaces ChatGPT and Claude for someone who wants agents that run on a schedule, work in groups and remember in a form you can edit, with a large tool library, on their own deployment. It is the agent-workspace pick on the open-source alternatives to ChatGPT page.
Where it falls short: the licence is source available. You may run it commercially unmodified, but building and distributing a changed version needs a paid licence, and LobeHub reserves the right to change the terms. The README names only OpenAI in its setup. It says it is under active development and that its plugin system is undergoing major work. For an MIT platform, LibreChat or AnythingLLM.
How does LobeHub run?
Deploy with a one-click button to Vercel, Zeabur, Sealos, RepoCloud or Alibaba Cloud, or on your own server with the setup script and docker compose up. Set a model API key; the README's example marks OPENAI_API_KEY as required. A hosted app runs at app.lobehub.com and the licence names a cloud edition; the README states no price.
Who is LobeHub for?
Someone who wants scheduled, collaborating agents with a large tool library, and who will run it as shipped. A team that needs to modify and redistribute it, or that wants Anthropic or local models named in the setup, should pick LibreChat or Open WebUI.
What limits does the README state?
From the README: "LobeHub is currently under active development"; "The plugin system is currently undergoing major development"; Vercel-assigned domains have DNS problems in some areas; one-click Vercel deploys may show constant update prompts.
Questions people ask
Is LobeHub open source?
Not by the OSI definition. The LobeHub Community License starts from Apache-2.0 and adds conditions: commercial use without modifying the source is allowed, but developing and distributing a derivative work requires a commercial licence, and contributors agree LobeHub may change the terms. This site labels it source available.
What happened to LobeChat?
The project is now LobeHub, and its repository moved from lobehub/lobe-chat to lobehub/lobehub; the old address redirects. The README presents it as an agent workspace rather than a chat interface, with agent groups, scheduled runs and editable memory. The licence file still refers to LobeChat by name.
Can I use LobeHub for my business?
The licence allows commercial use as a front end and back end without modifying the source code. If you change it and distribute the result, you need a commercial licence from LobeHub. Read the LICENSE file before building a product on it.
How does LobeHub compare with ChatGPT?
LobeHub focuses on agents you build, group and schedule, with a large plugin library and memory you can edit, on your own deployment. ChatGPT offers OpenAI's models and features as a service. The open-source alternatives to ChatGPT page compares it with LibreChat, Open WebUI and the desktop clients.
Which models does LobeHub support?
The README says it gives access to any model and any modality, but its setup only names OpenAI, marking the OpenAI key as required. Other providers are configured through the documentation rather than the README, so this page does not list them.
Sources
- LobeHub README and LobeHub Community License: github.com/lobehub/lobehub, read 2026-09-17.
Compared with the others
On the open-source alternatives to ChatGPT page, LobeHub is the pick for scheduled, collaborating agents. Agent builder, agent groups, scheduled runs, editable memory and a library of 10,000+ tools and MCP plugins.
Also on that page: Open WebUI for a private chatgpt for a whole team, LibreChat for every provider in one mit platform, AnythingLLM for chatting with your documents.
On the open-source alternatives to Claude page, LobeHub is the pick for scheduled, collaborating agents. Agent builder, agent groups, scheduled runs, editable memory and a library of 10,000+ tools and MCP plugins.
Also on that page: LibreChat for claude plus agents, artifacts and a code sandbox, Cherry Studio for many providers and 300+ assistants in one desktop app, Chatbox for one app across desktop and phone.
More chat interfaces
A self-hosted chat platform for Ollama and OpenAI-compatible models with document search over 9 vector databases, web search, image generation, roles and single sign-on, under a BSD-3 licence with an added branding clause.
An MIT all-in-one AI app from Mintplex Labs for chatting with your documents with citations, running agents with MCP, and using any of 39 model providers including local llama.cpp and Ollama, as a desktop app or a multi-user Docker server.
An AGPL-3.0 desktop client for Windows, macOS and Linux that connects OpenAI, Gemini, Anthropic and local Ollama or LM Studio models, with 300+ preset assistants, side-by-side model chats, file support and MCP servers.
An AGPL-3.0 desktop app for local models, formerly text-generation-webui, with five engines including llama.cpp and ExLlamaV3, an OpenAI and Anthropic-compatible API, tool calling, file chat, image generation and LoRA training, fully offline.
An Apache-2.0 desktop app from Menlo Research that downloads and runs open models from Hugging Face on your own computer, connects to OpenAI, Anthropic and others when you want, serves a local OpenAI-compatible API, and supports MCP.
Added September 17, 2026. Every claim above comes from the project's README, LICENSE or model card, read on September 17, 2026, or from the GitHub API on the date shown in the panel. Found an error? Write to reveneau@licheo.com and it is fixed in the next weekly pass. Repository: github.com/lobehub/lobehub.
Running one of these inside your own environment, with your own data and your own security rules, is the kind of work Reveneau does. Read how a forward deployed engagement works.