Welcome to wechat-chatgpt 👋

Use ChatGPT On Wechat via wechaty
English | 中文文档

🌟 Features

Interact with WeChat and ChatGPT:
- Use ChatGPT on WeChat with wechaty and Official API
- Add conversation support
- Support command setting
Deployment and configuration options:
- Add Dockerfile, deployable with docker
- Support deployment using docker compose
- Support Railway and Fly.io deployment
Other features:
- Support Dall·E
- Support whisper
- Support setting prompt
- Support proxy (in development)

🚀 Usage

Use with Railway(PaaS, Free, Stable, ✅Recommended)
Use with Fly.io(Paas, Free, ✅Recommended)
Use with docker(Self-hosted, Stable, ✅Recommended)
Use with docker compose(Self-hosted, Stable, ✅Recommended)
Use with nodejs(Self-hosted)

Use with Railway

Railway offers $5 or 500 hours of runtime per month

Click the Railway button to go to the Railway deployment page
Click the Deploy Now button to enter the Railway deployment page
Fill in the repository name and OPENAI_API_KEY (need to link GitHub account)
Click the Deploy button
Click the View Logs button and wait for the deployment to complete

Use with Fly.io

Please allocate 512MB memory for the application to meet the application requirements

fly.io offers free bills up to $5(Free Allowances 3 256MB are not included in the bill)

Install flyctl

 # macOS
 brew install flyctl
 # Windows
 scoop install flyctl
 # Linux
 curl https://fly.io/install.sh | sh

Clone the project and enter the project directory

git clone https://github.com/fuergaosi233/wechat-chatgpt.git && cd wechat-chatgpt

Create a new app

➜ flyctl launch 
 ? Would you like to copy its configuration to the new app? No
 ? App Name (leave blank to use an auto-generated name): <YOUR APP NAME>
 ? Select region: <YOUR CHOOSE REGION>
 ? Would you like to setup a Postgresql database now? No
 ? Would you like to deploy now? No

Configure the environment variables

flyctl secrets set OPENAI_API_KEY="<YOUR OPENAI API KEY>" MODEL="<CHATGPT-MODEL>"

Deploy the app
```
flyctl deploy
```

Use with docker

# pull image
docker pull holegots/wechat-chatgpt
# run container
docker run -d --name wechat-chatgpt \
    -e OPENAI_API_KEY=<YOUR OPENAI API KEY> \
    -e MODEL="gpt-3.5-turbo" \
    -e CHAT_PRIVATE_TRIGGER_KEYWORD="" \
    -v $(pwd)/data:/app/data/wechat-assistant.memory-card.json \
    holegots/wechat-chatgpt:latest
# View the QR code to log in to wechat
docker logs -f wechat-chatgpt

How to get OPENAI API KEY? Click here

Use with docker compose

# Copy the configuration file according to the template
cp .env.example .env
# Edit the configuration file
vim .env
# Start the container
docker-compose up -d
# View the QR code to log in to wechat
docker logs -f wechat-chatgpt

Use with nodejs

You need NodeJS 18.0.0 version and above

# Clone the project
git clone https://github.com/fuergaosi233/wechat-chatgpt.git && cd wechat-chatgpt
# Install dependencies
npm install
# Copy the configuration file according to the template
cp .env.example .env
# Edit the configuration file
vim .env
# Start project
npm run dev

Please make sure your WeChat account can log in WeChat on web

📝 Environment Variables

name	description
API	API endpoint of ChatGPT
OPENAI_API_KEY	create new secret key
MODEL	ID of the model to use. Currently, only gpt-3.5-turbo and gpt-3.5-turbo-0301 are supported.
TEMPERATURE	What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
CHAT_TRIGGER_RULE	Private chat triggering rules.
DISABLE_GROUP_MESSAGE	Prohibited to use ChatGPT in group chat.
CHAT_PRIVATE_TRIGGER_KEYWORD	Keyword to trigger ChatGPT reply in WeChat private chat
BLOCK_WORDS	Chat blocker words, (works for both private and group chats, Use, Split)
CHATGPT_BLOCK_WORDS	The blocked words returned by ChatGPT(works for both private and group chats, Use, Split)

📝 Using Custom ChatGPT API

https://github.com/fuergaosi233/openai-proxy

# Clone the project
git clone https://github.com/fuergaosi233/openai-proxy
# Install dependencies
npm install && npm install -g wrangler && npm run build
# Deploy to CloudFlare Workers
npm run deploy
# Custom domain (optional)
Add `Route` to `wrangler.toml`
routes = [
    { pattern = "Your Custom Domain", custom_domain = true },
]

⌨️ Commands

Enter in the WeChat chat box

/cmd help # Show help
/cmd prompt <PROMPT> # Set prompt
/cmd clear # Clear all sessions since last boot

✨ Contributor

🤝 Contributing

Contributions, issues and feature requests are welcome!
Feel free to check issues page.

Show your support

Give a ⭐️ if this project helped you!

chenjoya/chatbot