vual/ChatGPT-Next-Web-Pro
基于chatgpt-next-web,增加了midjourney绘画功能,支持mj-plus的ai换脸和局部重绘,接入了stable-diffusion,支持oss,支持接入fastgpt知识库,支持suno,支持luma。支持dall-e-3、gpt-4-vision-preview、whisper、tts等多模态模型,支持gpt-4-all,支持GPTs商店。新增加了支持后台管理的版本,包括登录注册,平台模型apikey管理,套餐管理,消息保存等。
ChatGPT‑Next‑Web‑Pro
What it is – An open‑source, self‑hosted web front‑end for OpenAI‑style large language models that extends the popular ChatGPT‑Next‑Web UI with a large set of “pro” features:
- multimodal support (image OCR, audio transcription, text‑to‑speech, vision)
- integrated generation of images, music and video (Midjourney, Stable‑Diffusion, DALL‑E‑3, Suno, Luma, etc.)
- optional backend services for user accounts, subscription/plan management, file storage, payment gateways and admin panels.
- works in two flavours:
- “No‑backend” – pure front‑end that only needs an OpenAI (or compatible) API key.
- “With‑backend” – full SaaS‑style stack with login, API‑key management, usage‑based plans, and a management UI.
Key Features (as listed in the README)
| Category | Feature | No‑backend (v3.8) | No‑backend (v3.9+) | With‑backend |
|---|---|---|---|---|
| Core | Full ChatGPT‑Next‑Web UI | ✅ | ✅ | ✅ |
| File handling | Parse uploaded PDFs, Word, PPT, Excel, images (OCR), audio, HTML, txt, zip | ❌ | ❌ | ✅ |
| Storage | S3‑compatible storage (Aliyun OSS, Tencent COS, MinIO, AWS, Cloudflare R2, custom) | ❌ | ✅ | ✅ |
| Drawing | Midjourney (standard & plus), independent drawing panel, Stable‑Diffusion (sd‑webui), DALL‑E‑3, gpt‑image‑1 | ✅ | ✅ | ✅ |
| Music | Suno (chat‑style) | ✅ | ✅ | ✅ |
| Video | Luma | ❌ | ✅ | ✅ |
| Multimodal | Image recognition, Whisper transcription, TTS | ✅ | ✅ | ✅ |
| Model extensions | Reverse‑engineered models (gpts, gpt‑4o‑all) | ✅ | ✅ | ✅ |
| Knowledge base | FastGPT knowledge‑base as a model | ✅ | ✅ | ✅ |
| UI tweaks | Custom site title, subtitle, icon; token‑count display; theme support; custom avatar/name; plugin button customization | ✅ | ✅ | ✅ |
| Auth (with‑backend only) | Phone, email, WeChat QR login | ❌ | ❌ | ✅ |
| Admin panel (with‑backend) | Platform, model, API‑key, plan, order, member, conversation, payment (WeChat, EasyPay, HuPiJiao), sensitive‑word and quick‑prompt management | ❌ | ❌ | ✅ |
Getting Started
1️⃣ No‑backend (quick Docker run)
# Pull the image (example version 3.9.18)
docker pull registry.cn-hangzhou.aliyuncs.com/ann-chat/chatgpt-next-web-pro:3.9.18
# Run it – replace the placeholder key with your own OpenAI key
docker run -d -p 3000:3000 \
-e OPENAI_API_KEY="sk-xxxxxx" \
registry.cn-hangzhou.aliyuncs.com/ann-chat/chatgpt-next-web-pro:3.9.18
- The container always listens on internal port 3000; map it to any host port you like (
-p 3030:3000). - If you need to route requests through a proxy or a self‑hosted OpenAI‑compatible endpoint, add
-e BASE_URL="https://your‑proxy". - Additional environment variables can be added with
-e NAME=VALUE.
2️⃣ With‑backend (full SaaS stack)
# Grab the compose file (or copy it manually if the curl returns HTML)
curl -o docker-compose.yml \
https://github.com/vual/ChatGPT-Next-Web-Pro/tree/main/docker/with-backend/docker-compose.yml
# Edit the file – set your OpenAI key, storage credentials, payment keys, etc.
# Then pull images and start the services
docker compose pull
docker compose up -d # omit -d first time to watch logs
- After launch, the user portal is at
https://nextweb.annyun.cn(supports WeChat QR, email registration). - The admin console is at
https://nextadmin.annyun.cn(default credentialsadmin / 123456; change immediately). - Nginx reverse‑proxy configuration is provided in
docs/nginx.conf.
Who Might Use It?
- Developers / hobbyists who want a locally hosted ChatGPT‑style UI with extra media generation capabilities without building everything from scratch.
- Small teams or startups that need a quick‑to‑deploy internal AI assistant with user accounts, usage tracking, and optional paid plans.
- Educators / researchers looking for a sandbox to experiment with multimodal prompts (image‑to‑text, audio‑to‑text, image generation, music, video) on top of OpenAI‑compatible APIs.
License & Security
- The repository states that the code is fully open‑source and “security is identical to the original ChatGPT‑Next‑Web”.
- Authors explicitly claim no malicious code and encourage users to sniff network traffic to verify.
- No explicit license is mentioned in the README; you should check the repository’s
LICENSEfile for the exact terms before commercial use.
Where to Get Help
- Documentation – detailed feature lists, parameter explanations, deployment steps, and FAQ are in the
docs/folder. - Community – a WeChat group QR code is shown in the README; the author’s contact WeChat (
822784588) is provided for source‑code requests. - Related projects –
lobe-chat-pro(stronger drawing/music/video panel) andaiiai-api(extended model support) are linked for further extensions.
Bottom Line
ChatGPT‑Next‑Web‑Pro is a genuine, feature‑rich fork of a popular open‑source ChatGPT UI, adding multimodal generation, file handling, storage back‑ends, and an optional full‑stack admin system. It can be run instantly with a single Docker command for a simple key‑only setup, or deployed as a complete SaaS‑like service with user management and payment integration.
Related
- Project
- Project
- Dispatch
- Project
- Dispatch