chatgpt-demo
FreeMinimal web UI for ChatGPT.
FreeFree tier
Inputs: textOutputs: text
About chatgpt-demo
A minimal open-source web UI for interacting with OpenAI's GPT-3.5 Turbo API. Built with Astro and React, it provides a simple chat interface that can be self-hosted on Vercel, Netlify, Docker, or Sealos. The project is designed to be lightweight and easy to configure via environment variables (API key, proxy, model selection, site password). Originally created as a demo, it now directs users to 'Anse' for more advanced features like multi-chat and image generation.
Key Features
Minimal chat interface for OpenAI GPT-3.5 Turbo API
Self-hosted with easy deployment via Vercel, Netlify, Docker, or Sealos
Configurable via environment variables (API key, proxy, model, site password)
Built with Astro and React
Open source under MIT license
Live preview available (currently limited due to exhausted API key)
Pros & Cons
Pros
- Lightweight and minimal — easy to understand and modify
- Multiple deployment options (Vercel, Netlify, Docker, Sealos)
- Open source with active community (8k stars on GitHub)
- Simple configuration via environment variables
- Fully controllable — no vendor lock-in for the frontend
Cons
- Requires own OpenAI API key (usage costs not included)
- Demo site currently unavailable due to exhausted API key
- Limited to GPT-3.5 Turbo; no support for GPT-4 or image generation
- Basic single-chat interface; multi-chat and advanced features require migrating to Anse
- Setup requires technical knowledge (Node, pnpm, API key management)
Best For
Self-hosted ChatGPT frontend for personal or internal useEducational demo for interacting with GPT-3.5 TurboCustom chatbot UI that can be deployed quicklyBase for building more advanced chat applications (with Anse)
FAQ
What is chatgpt-demo?
A minimal open-source web UI for interacting with OpenAI's GPT-3.5 Turbo API. It provides a simple chat interface that can be self-hosted.
How do I run chatgpt-demo locally?
Ensure Node v18+ and pnpm are installed. Copy .env.example to .env, add your OpenAI API key (OPENAI_API_KEY), then run 'pnpm install' and 'pnpm run dev'. The app runs on http://localhost:3000.
How can I deploy chatgpt-demo?
The project can be deployed on Vercel, Netlify, Docker, or Sealos. For Vercel, fork the repo and import with environment variables. For Netlify, fork and connect via Netlify drag-and-drop. Docker images are available at ddiu8081/chatgpt-demo. Sealos deployment via App Launchpad is also supported.
What environment variables are available?
OPENAI_API_KEY (required), HTTPS_PROXY, OPENAI_API_BASE_URL, HEAD_SCRIPTS, PUBLIC_SECRET_KEY, SITE_PASSWORD, and OPENAI_API_MODEL.
Is the demo site working?
As of the repository notice, the official demo (https://chatgpt.ddiu.me) is not available because the API key limit has been exhausted. Users are encouraged to self-host.
Does this support multi-chat or image generation?
No, this demo is limited to a single chat session with GPT-3.5 Turbo. For multi-chat, image generation, and more powerful features, the project recommends using 'Anse' (linked in the README).