LibreChat
Anthropic is first in its provider list, alongside Bedrock and Vertex, with agents, MCP, artifacts, a code sandbox for eight languages and sign-in, under MIT.
Know firstA server to run, with no desktop or mobile client.
Chat interfaces6 projects, checked September 17, 2026
Open-source alternatives to Claude
The open-source alternatives to the Claude app are LibreChat, Open WebUI, AnythingLLM, LobeHub, Chatbox and Cherry Studio. Most of them can still use Claude models through your own Anthropic key, next to OpenAI, Google and local models, so you choose the interface and the model separately. LibreChat and Cherry Studio name Anthropic directly; Open WebUI reaches it through an OpenAI-compatible gateway. A team switches to own the interface, add its own documents and sign-in, or mix Claude with other models.
- Claude plus agents, artifacts and a code sandbox
- LibreChat
- Many providers and 300+ assistants in one desktop app
- Cherry Studio
- One app across desktop and phone
- Chatbox
4 of 6 carry an OSI licence. 6 of 6 run on your own machines. Figures read from GitHub on September 17, 2026.
Pick by need
One project per need, with the reason from its own README and the one thing to know before you choose it.
Cherry Studio
AGPL-3.0 client with OpenAI, Gemini, Anthropic, Ollama and LM Studio, 300+ preset assistants, side-by-side model chats, file support and MCP.
Know firstSpeech, OCR and mobile apps are still on its roadmap.
Chatbox
GPL-3.0 client for OpenAI, Claude, Gemini and local Ollama models on Windows, macOS, Linux, iOS and Android, with conversations stored on the device.
Know firstNo document chat, web search or MCP described; a separate pro edition exists.
AnythingLLM
MIT, with document chat and source citations, agents with MCP, and 39 model providers including local llama.cpp and Ollama, as a desktop app or a Docker server.
Know firstMulti-user support is Docker only, and telemetry is on until you turn it off.
Open WebUI
Documents over 9 vector databases, web search, image generation, voice, and roles with LDAP, OAuth single sign-on and SCIM, on your own server with Ollama or any OpenAI-compatible model.
Know firstSource available: above 50 users you may not remove its branding without permission or a paid licence.
LobeHub
Agent builder, agent groups, scheduled runs, editable memory and a library of 10,000+ tools and MCP plugins.
Know firstSource available: distributing a modified version needs a paid licence, and the README names only OpenAI in setup.
What to take away
- Most alternatives keep Claude as a model option and replace only the interface, so you can mix Claude with OpenAI, Google and local models.
- LibreChat is the closest to the Claude app's feature set, with artifacts and a code sandbox, under MIT.
- The open apps need an Anthropic API key; a Claude subscription does not carry over.
- The Claude app still wins on Anthropic's own features built around its models, with nothing to run.
The projects, side by side
The facts that decide most choices. Each project's page carries the full panel.
| Project | Best for | Openness | Licence | Runs as | Own machines | Stars |
|---|---|---|---|---|---|---|
| LibreChat | Claude plus agents, artifacts and a code sandbox | OSI | MIT | Docker, Docker Compose | Yes | 44,252 |
| Cherry Studio | Many providers and 300+ assistants in one desktop app | OSI | AGPL-3.0 | Desktop app | Yes | 51,940 |
| Chatbox | One app across desktop and phone | OSI | GPL-3.0 | Desktop app, Mobile app | Yes | 41,795 |
| AnythingLLM | Chatting with your documents | OSI | MIT | Desktop app, Docker, Bare metal | Yes | 66,155 |
| Open WebUI | A private ChatGPT for a whole team | Source available | See LICENSE | pip, Docker, Kubernetes | Yes | 152,426 |
| LobeHub | Scheduled, collaborating agents | Source available | See LICENSE | Docker, One-click deploy (Vercel, Zeabur, Sealos and others) | Yes | 82,574 |
Licence and star figures read from GitHub on September 17, 2026. Stars are shown as a dated fact and were not used to rank this page.
Each project, in turn
An MIT self-hosted chat platform that puts Anthropic, OpenAI, Google, Bedrock, Azure, Ollama and any OpenAI-compatible model in one interface, with agents, MCP, a code sandbox, web search, artifacts and multi-user sign-in.
Anthropic is first in its provider list, alongside Bedrock and Vertex, with agents, MCP, artifacts, a code sandbox for eight languages and sign-in, under MIT.
Know firstA server to run, with no desktop or mobile client.
- Licence
- MIT
- Runs as
- Docker, Docker Compose
- Platforms
- Self-hosted web app
- Hosted option
- None from the maintainers
- Last release
- No tagged release
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.
AGPL-3.0 client with OpenAI, Gemini, Anthropic, Ollama and LM Studio, 300+ preset assistants, side-by-side model chats, file support and MCP.
Know firstSpeech, OCR and mobile apps are still on its roadmap.
- Licence
- AGPL-3.0
- Runs as
- Desktop app
- Platforms
- Windows, macOS, Linux
- Hosted option
- None from the maintainers
- Last release
- v2.0.14, Sep 9, 2026
The GPL-3.0 community edition of a desktop and mobile chat client for ChatGPT, Claude, Gemini and local Ollama models that keeps conversations on your device, with a separate pro edition synced from the same code.
GPL-3.0 client for OpenAI, Claude, Gemini and local Ollama models on Windows, macOS, Linux, iOS and Android, with conversations stored on the device.
Know firstNo document chat, web search or MCP described; a separate pro edition exists.
- Licence
- GPL-3.0
- Runs as
- Desktop app, Mobile app
- Platforms
- Windows 10, macOS 11 or later, Linux (AppImage), iOS, Android, Web
- Hosted option
- Sold by the maintainers
- Last release
- v1.23.3, Sep 16, 2026
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.
MIT, with document chat and source citations, agents with MCP, and 39 model providers including local llama.cpp and Ollama, as a desktop app or a Docker server.
Know firstMulti-user support is Docker only, and telemetry is on until you turn it off.
- Licence
- MIT
- Runs as
- Desktop app, Docker, Bare metal
- Platforms
- macOS, Windows, Linux, Docker, Mobile app (separate repository)
- Hosted option
- Sold by the maintainers
- Last release
- v1.16.1, Aug 27, 2026
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.
Documents over 9 vector databases, web search, image generation, voice, and roles with LDAP, OAuth single sign-on and SCIM, on your own server with Ollama or any OpenAI-compatible model.
Know firstSource available: above 50 users you may not remove its branding without permission or a paid licence.
- Licence
- See the LICENSE file
- Runs as
- pip, Docker, Kubernetes
- Platforms
- Any browser, including an installable web app on phones, Server on Docker, Kubernetes or Python 3.11
- Hosted option
- None from the maintainers
- Last release
- v0.11.3, Aug 31, 2026
LobeHub, formerly LobeChat, is an agent workspace with an agent builder, agent groups, scheduled runs, editable memory and a library of over 10,000 tools and MCP plugins, under a community licence that needs a paid licence to distribute a modified version.
Agent builder, agent groups, scheduled runs, editable memory and a library of 10,000+ tools and MCP plugins.
Know firstSource available: distributing a modified version needs a paid licence, and the README names only OpenAI in setup.
- Licence
- See the LICENSE file
- Runs as
- Docker, One-click deploy (Vercel, Zeabur, Sealos and others)
- Platforms
- Self-hosted web app
- Hosted option
- Sold by the maintainers
- Last release
- v2.2.17, Sep 11, 2026
How to choose
The questions that settle it, in the order they usually come up.
Do you want to keep using Claude models?
Yes, with an Anthropic key: LibreChat, Cherry Studio, Chatbox and AnythingLLM name it directly; Open WebUI needs a gateway such as OpenRouter.
A desktop app or a server?
Desktop: Cherry Studio, Chatbox or AnythingLLM. Server for a team: LibreChat, Open WebUI or LobeHub.
Do you need artifacts or code execution?
LibreChat renders artifacts and runs code in a sandbox. The others render Markdown and code only.
Where Claude is still the better choice
Claude, in the vendor's own words, is "a next generation AI assistant built by Anthropic and trained to be safe, accurate, and secure to help you do your best work", by Anthropic's own description. claude.com. The cases below are where that still wins.
- You want Anthropic's own features, projects, artifacts and the rest, designed around Claude, with nothing to run.
- You want the same assistant on every device through one Anthropic account.
- You use Claude through a subscription rather than an API key; the open apps mostly need a key.
- You want Anthropic accountable for the whole product.
How this page was checked
Numbers from GitHub, dated
Stars, licence, last commit and last release are read from the GitHub API by a script every week and carry the date they were read. Nobody types them.
Claims from the README
What each project does is written from its own README and LICENSE, read in full on the date shown, and phrased as the project's claim. The sources are listed at the end.
What is never said
That a project is faster, better or cheaper than Claude. A price for anyone. A user count. A roadmap. If a fact is not on this page, it was not in the source.
- OSI Open source (OSI licence)
- The code, and the weights where the project is a model, carry a licence on the OSI approved list.
- Source available Source available
- The code is public, and the licence limits what you may do with it. Read the LICENSE file before commercial use.
- Open weights Open weights
- The model weights can be downloaded and run. The training data or code is closed, or the weights carry a restriction the code does not.
Questions people ask
What is the best open-source alternative to the Claude app?
LibreChat (MIT), which lists Anthropic first among its providers and adds agents, MCP, a code sandbox and artifacts, on your own server. For a desktop app, Cherry Studio (AGPL-3.0) or Chatbox (GPL-3.0), both connecting Claude with your key. For a team with documents and roles, Open WebUI.
Can I use Claude models in an open-source chat app?
Yes, with an Anthropic API key. LibreChat, AnythingLLM, Chatbox and Cherry Studio list Anthropic or Claude directly, and LibreChat also reaches Claude through AWS Bedrock and Vertex AI. Open WebUI's README names OpenAI-compatible APIs, so Claude needs a gateway such as OpenRouter in front.
Do these apps support artifacts like Claude?
LibreChat renders artifacts, React, HTML and Mermaid inside the chat, and runs code in a sandbox for eight languages. Cherry Studio renders Mermaid charts and code. The other apps render Markdown and code but do not describe artifacts in their READMEs.
Do these alternatives support MCP?
Yes, most of them: LibreChat's agents use MCP servers, Open WebUI connects MCP, MCPO and OpenAPI tool servers, AnythingLLM is MCP-compatible, LobeHub counts over 10,000 tools and MCP plugins, and Cherry Studio supports MCP servers. Chatbox's README does not mention MCP.
Are these Claude alternatives free?
The software is free to run. The model costs what Anthropic or another provider charges for API use, or nothing beyond your hardware if you run a local model through Ollama. Open WebUI and LobeHub have licence conditions described on their own pages.
Sources
- claude.com, Anthropic's own description, read September 17, 2026
- LibreChat README, read September 17, 2026
- Cherry Studio README, read September 17, 2026
- Chatbox README, read September 17, 2026
- Open WebUI README, read September 17, 2026
- AnythingLLM README, read September 17, 2026
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.