Web UI for ChatGPT, optionally with Midjourney image generation
Top 71.0% on sourcepulse
This project provides a well-designed, one-click deployable ChatGPT web UI, targeting users who want a private, customizable ChatGPT interface. It offers features like Markdown support, prompt templates, and local data storage, enhancing user experience and privacy.
How It Works
The UI is built with a focus on ease of deployment and user experience, leveraging Vercel for one-click deployment. It supports streaming responses, Markdown rendering (including LaTeX and Mermaid), and PWA capabilities. A key architectural choice is storing all data locally in the browser, prioritizing user privacy. It also includes features for creating, sharing, and debugging chat tools via prompt templates.
Quick Start & Requirements
CODE
(access password), BASE_URL
(for custom OpenAI endpoints), OPENAI_ORG_ID
, HIDE_USER_API_KEY
, and DISABLE_GPT4
. For Midjourney integration, MIDJOURNEY_API_KEY
, MIDJOURNEY_API_URL
, and MIDJOURNEY_IMG_PROXY
are needed.docker pull zo3i/chatgptwithmidjourney
and docker run -d -p 3000:3000 -e OPENAI_API_KEY="sk-xxxx" -e CODE="your-password" zo3i/chatgptwithmidjourney
.Highlighted Details
Maintenance & Community
Licensing & Compatibility
Limitations & Caveats
The project explicitly states that user login, account management, and cloud sync are not planned features. Automatic updates require forking the repository and enabling GitHub Actions workflows due to Vercel's deployment model.
2 years ago
1 day