NextChat

license

MIT(core, public source)

enterprise-tier

NextChat Enterprise (permission controls, team resource mgmt, internal knowledge base/RAG)

source

public(the app IS the repo

core features withheld

yes — Enterprise Edition withholds the admin panel, member/resource/knowledge-base permission control, internal knowledge-base integration and security auditing from the MIT repo

An OSI MIT core with a separate proprietary Enterprise Edition holding multi-user permissions and internal knowledge-base RAG, which makes this open core rather than fully open, on the same reading applied to LiteLLM.

• https://api.github.com/repos/ChatGPTNextWeb/NextChat/contents/ recorded 2026-08-11

  Repository root listing. The repo license endpoint returns MIT, and the tree is the whole application: app, src-tauri, public, docs, plus Dockerfile, docker-compose.yml, vercel.json and .env.template. There is no `ee/` or `enterprise/` directory, so the Enterprise Edition is not in this repository. The README's deploy buttons for Vercel, Zeabur and Gitpod run this same source.

• https://cdn.jsdelivr.net/gh/ChatGPTNextWeb/NextChat@main/LICENSE recorded 2026-09-29

  The repository LICENSE file opens "MIT License" with "Copyright (c) 2023-2025 NextChat" and then the standard MIT permission grant, so the published source is under MIT.

• https://cdn.jsdelivr.net/gh/ChatGPTNextWeb/NextChat@main/README.md recorded 2026-09-29

  The README lists deployment by one-click Vercel and Zeabur buttons and a "Compact client (~5MB) on Linux/Windows/MacOS", so the app is the repository itself. Its Enterprise Edition section describes "Permission Control" as "member permissions, resource permissions, and knowledge base permissions, all controlled via a corporate-grade Admin Panel", plus "Security Auditing" and "Knowledge Integration". The Features list describes the opposite design: "Privacy first, all data is stored locally in the browser". Nothing in the README places the admin panel or permission model in the MIT repository, and the roadmap leaves "local knowledge base" unchecked.

• https://github.com/ChatGPTNextWeb/NextChat recorded 2026-08-13

  MIT license; cross-platform chat client; Enterprise Edition (permissions, team mgmt, internal knowledge base)

• https://github.com/ChatGPTNextWeb/NextChat/blob/main/README.md recorded 2026-08-13

  The README's own Enterprise Edition section, quoted rather than paraphrased, lists what the paid edition adds - Brand Customization, Resource Integration described as "unified configuration and management of dozens of AI resources by company administrators", Permission Control described as "clearly defined member permissions, resource permissions, and knowledge base permissions, all controlled via a corporate-grade Admin Panel", Knowledge Integration "combining your internal knowledge base with AI capabilities", Security Auditing, Private Deployment, and Continuous Updates. Contact is business@nextchat.club. None of that admin panel, permission model or knowledge base exists in the MIT repository, whose own Features list describes the opposite design: "Privacy first, all data is stored locally in the browser". So this is functionality withheld from the published source rather than a hosted service beside it, which is the gated reading.

NextChat ships as a one-click Vercel deploy alongside desktop and mobile builds, but no download or monthly-active-user figure is published for it, leaving GitHub stars as the only adoption signal. A star is not a use, so easy deployment does not translate into a measurable count of actual use.

• https://github.com/ChatGPTNextWeb/NextChat recorded 2026-08-13

  88.2k stars; v2.16.1 latest release; cross-platform deployment

Solid lightweight cross-platform client with plugins/MCP/vision, but RAG and multi-user are gated to Enterprise; thinner OSS feature surface than Open WebUI/LibreChat/AnythingLLM.

• https://cdn.jsdelivr.net/gh/ChatGPTNextWeb/NextChat@main/README.md recorded 2026-09-29

  The README describes NextChat "with Claude, DeepSeek, GPT4 & Gemini Pro support" and lists environment settings for Azure, Baidu, ByteDance and Alibaba endpoints. The roadmap checks off "Artifacts", "Plugins: support network search, calculator, any other apis etc." and "Supports Realtime Chat", and it documents "ENABLE_MCP" for the Model Context Protocol and "VISION_MODELS" for vision-capable models. Knowledge-base and permission features appear only under the Enterprise Edition heading.

• https://github.com/ChatGPTNextWeb/NextChat recorded 2026-08-13

  multi-provider, plugins, vision, MCP, artifacts; RAG/multi-user in Enterprise Edition