ChatGPT UI
[English](./README.md) | [ไธญๆ](./docs/zh/README.md)
A ChatGPT web client that supports multiple users, multiple database connections for persistent data storage, supports i18n. Provides Docker images and quick deployment scripts.
https://user-images.githubusercontent.com/46235412/227156264-ca17ab17-999b-414f-ab06-3f75b5235bfe.mp4
## ๐ขUpdates
2023-03-27
๐ Support gpt-4 model. You can select the model in the "Model Parameters" of the front-end.
The GPT-4 model requires whitelist access from OpenAI.
2023-03-23
Added web search capability to generate more relevant and up-to-date answers from ChatGPT!
This feature is off by default, you can turn it on in `Chat->Settings` in the admin panel, there is a record `open_web_search` in Settings, set its value to True.
2023-03-15
Add "open_registration" setting option in the admin panel to control whether user registration is enabled. You can log in to the admin panel and find this setting option under `Chat->Setting`. The default value of this setting is `True` (allow user registration). If you do not need it, please change it to `False`.
2023-03-04
**Update to the latest official chat model** `gpt-3.5-turbo`
**๐๐๐Provide a shell script that can be used to quickly deploy the service to server** [Quick start](#one-click-depolyment)
2023-02-24
Version 2 is a major update that separates the backend functionality as an independent project, hosted at [chatgpt-ui-server](https://github.com/WongSaang/chatgpt-ui-server).
If you still wish to use the old version, please visit the [v1 branch](https://github.com/WongSaang/chatgpt-ui/tree/v1).
## Version 2 introduces the following new features:
- ๐ Separation of the frontend and backend, with the backend now using the Python-based Django framework.
- ๐ User authentication, supporting multiple users.
- ๐ Ability to store data in an external database (defaulting to Sqlite).
- ๐ Session persistence, allowing the API to answer questions based on your context.
## ๐ One-click deployment
Note: This script has only been tested on Ubuntu Server 22.04 LTS.
```bash
bash <(curl -Ls https://raw.githubusercontent.com/WongSaang/chatgpt-ui/main/deployment.sh)
```
> If you have a domain name, you can point it to the server's IP address using DNS resolution. Of course, using the server's IP address directly is also possible.
> During the script's execution, you will be prompted to enter a domain name. If you do not have a domain name, you can enter the server's IP address directly.
### After the deployment is complete
Access `http(s)://your.domain:9000/admin` / IP `http(s)://123.123.123.123:9000/admin` to log in to the administration panel.
Default superuser: `admin`
Default password: `password`
Before you can start chatting, you need to add an OpenAI API key. In the Settings model, add a record with the name `openai_api_key` and the value as your API key.
Now you can access the web client at `http(s)://your.domain` or `http://123.123.123.123` to start chatting.
๐๐๐ Enjoy it!
## Quick start with Docker Compose
### Run services
Below is a docker-compose.yml template:
```yaml
version: '3'
services:
client:
image: wongsaang/chatgpt-ui-client:latest
environment:
- SERVER_DOMAIN=http://backend-web-server
depends_on:
- backend-web-server
ports:
- '80:80'
networks:
- chatgpt_ui_network
backend-wsgi-server:
image: wongsaang/chatgpt-ui-wsgi-server:latest
environment:
- APP_DOMAIN=${APP_DOMAIN:-localhost:9000} # CSRF whitelist๏ผAdd the address of your chatgpt-ui-web-server here, default is localhost:9000
- SERVER_WORKERS=3 # Number of gunicorn workers, default is 3
#- DB_URL=postgres://postgres:postgrespw@localhost:49153/chatgpt # If this parameter is not set, the built-in Sqlite will be used by default. It should be noted that if you do not connect to an external database, the data will be lost after the container is destroyed.
#- OPENAI_API_PROXY=https://openai.proxy.com/v1 # Proxy for https://api.openai.com/v1
- DJANGO_SUPERUSER_USERNAME=admin # default superuser name
- DJANGO_SUPERUSER_PASSWORD=password # default superuser password
- DJANGO_SUPERUSER_EMAIL=admin@example.com # default superuser email
- ACCOUNT_EMAIL_VERIFICATION=none # Determines the e-mail verification method during signup โ choose one of "none", "optional", or "mandatory". Default is "optional". If you don't need to verify the email, you can set it to "none".
# If you want to use the email verification function, you need to configure the following parameters
# - EMAIL_HOST=SMTP server address
# - EMAIL_PORT=SMTP server port
# - EMAIL_HOST_USER=
# - EMAIL_HOST_PASSWORD=
# - EMAIL_USE_TLS=True
# - EMAIL_FROM=no-reply@example.com #Default sender email address
ports:
- '8000:8000'
networks:
- chatgpt_ui_network
backend-web-server:
image: wongsaang/chatgpt-ui-web-server:latest
environment:
- BACKEND_URL=http://backend-wsgi-server:8000
ports:
- '9000:80'
depends_on:
- backend-wsgi-server
networks:
- chatgpt_ui_network
networks:
chatgpt_ui_network:
driver: bridge
```
### DB_URL schema
| Engine | URL |
|----------------------|--------------------------------------------------|
| PostgreSQL | ``postgres://USER:PASSWORD@HOST:PORT/NAME`` |
| MySQL | ``mysql://USER:PASSWORD@HOST:PORT/NAME`` |
| SQLite | ``sqlite:///PATH`` |
### Set API key
Access `http(s)://your.domain:9000/admin` / IP `http(s)://123.123.123.123:9000/admin` to log in to the administration panel.
Default superuser: `admin`
Default password: `password`
Before you can start chatting, you need to add an OpenAI API key. In the Settings model, add a record with the name `openai_api_key` and the value as your API key.
Now you can access the web client at `http(s)://your.domain` or `http://123.123.123.123` to start chatting.
## Donation
> If it is helpful to you, it is also helping me.
If you want to support me, Buy me a coffee โค๏ธ [https://www.buymeacoffee.com/WongSaang](https://www.buymeacoffee.com/WongSaang)
## Development
### Setup
Make sure to install the dependencies:
```bash
# yarn
yarn install
```
### Development Server
Start the development server on http://localhost:3000
```bash
yarn dev
```
### Production
Build the application for production:
```bash
yarn build
```