Open source

Open-source chat interfaces

Apps and self-hosted platforms that give you a ChatGPT-style chat window over models you choose: desktop clients for one person, and servers with sign-in, roles and document search for a team. Most connect to local models through Ollama as well as to hosted providers with your own keys.

ProjectReplacesOpennessStarsLast releaseSelf-host
Open WebUI

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.

Source available
152,426
Aug 31, 2026
Yes
LobeHub

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.

Source available
82,574
Sep 11, 2026
Yes
AnythingLLM

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.

OSI
66,155
Aug 27, 2026
Yes
Cherry Studio

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.

OSI
51,940
Sep 9, 2026
Yes
TextGen

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.

OSI
47,683
May 20, 2026
Yes
Jan

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.

OSI
44,516
Jul 23, 2026
Yes
LibreChat

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.

OSI
44,252
No tagged release
Yes
Chatbox

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.

OSI
41,795
Sep 16, 2026
Yes
Chat UI

Hugging Face's Apache-2.0 chat interface, the code behind HuggingChat, that talks to any OpenAI-compatible API, stores chats in MongoDB, calls MCP tools, and can route each message to a model automatically.

OSI
10,953
May 11, 2026
Yes

Questions people ask

What is the best open-source ChatGPT alternative?

For a team on a server: Open WebUI, the most complete, under a BSD-3 licence with a branding clause, or LibreChat (MIT) with Anthropic, OpenAI, Google and Ollama built in. For one person on a desktop: Jan (Apache-2.0) for local models, AnythingLLM (MIT) for chatting with documents, or Chatbox (GPL-3.0) across desktop and phone.

Which open-source chat apps run fully offline?

Jan, TextGen, AnythingLLM, Open WebUI, Cherry Studio and Chatbox all run local models, through their own llama.cpp engine or through Ollama and LM Studio. TextGen's README states zero telemetry and no remote update requests; AnythingLLM's telemetry is on by default until you turn it off.

Which open-source chat interfaces support multiple users?

Open WebUI (roles, groups, LDAP, OAuth single sign-on and SCIM), LibreChat (OAuth2, LDAP and email sign-in with an admin panel), AnythingLLM in its Docker version, and TextGen's multi-user mode for small trusted teams. The desktop apps are for one person.

Are these chat interfaces really open source?

Most are, under MIT, Apache-2.0, GPL-3.0 or AGPL-3.0. Two are source available: Open WebUI's licence forbids removing its branding above 50 users without permission or a paid licence, and LobeHub's needs a paid licence to distribute a modified version. Each project page states its licence and why.

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.