DDG-Chat: Deploying Chat2API Backend with One Click, DuckDuckGo to APIs

General Introduction

DDG-Chat is an open source project that aims to provide a one-click deployment to multiple platforms of the ChatGPT API backend. The project supports multiple models, including GPT-4o mini, Claude 3 Haiku, Llama 3.1 70B, and Mixtral 8x7B, all anonymously provided by DuckDuckGo. All models are provided anonymously by DuckDuckGo. Users can easily deploy and use the API backend through Vercel, Cloudflare Workers, Docker and other platforms.

Since the DDG API limits the number of concurrency to a single IP, it is recommended that you use Vercel for deployment, or if you use a local deployment such as Docker, make sure the project is running in a proxy pool.

Function List

• Supports multiple ChatGPT models
• One-click deployment to Vercel, Cloudflare Workers, Docker, etc.
• Provide API interface for third-party applications
• Supports multiple deployment methods, including cloud and local deployment
• Provide detailed deployment and usage documentation

Using Help

Installation and Deployment

Vercel deployment

1. Cloud Fork repository deployment: (Vercel One-Click Deployment)
   • Fork this repository to your GitHub account.
   • Go to the Vercel New Project page and import the repository you just Forked.
   • Click Deploy to complete the deployment.
2. Local Clone repository deployment:
   • Make sure you have a Node.js environment.
   • Execute the following command:

     npm i -g vercel
     vercel login
     git clone https://github.com/leafmoes/DDG-Chat.git ddg-chat
     cd ddg-chat
     npm run publish
     

Cloudflare Workers Deployment

1. Go to the Cloudflare Workers and Pages console and create a worker.
2. In the Workers settings, set the runtime to nodejs_compat.
3. Paste the code from the repositoryGo to your Workers and click Deploy.

Docker Deployment

1. Command line build:

   docker run -it -d --name ddg-chat -p 8787:8787 ghcr.io/leafmoes/ddg-chat:latest
   

2. Build using the docker-compose.yml file:
   • Download and save the docker-compose.yml file.
   • Start the service by running the following command in the directory where the file is located:

     docker-compose up -d
     

environment variable

Prefix address for # API calls
API_PREFIX = '/'
API Key that # uses as authentication for API calls
API_KEY = 'dummy_key'
# Number of retries for failed requests to the DDG
MAX_RETRY_COUNT = 3
# retry delay for failed requests to the DDG in ms
RETRY_DELAY = 5000

Using the API

invoke an interface

Use a third-party ChatGPT application to invoke the interface, such as ChatNextWeb. the following is a sample dialog request:

curl --request POST 'https://你的域名/v1/chat/completions' \
--header 'Content-Type: application/json' \
--data '{
"messages": [
{
"role": "user", {
"content": "Hello!"
}
],
"model": "gpt-4o-mini",
"stream": true
}'

Model Query

interviews https://你的域名/v1/models Allows you to view the currently supported models.

common problems

• Vercel deployment limitations: The free version takes 60 seconds for a single API request and 100,000 API calls per month.
• Error 429 ERR_SERVICE_UNAVAILABLE: Non-Vercel are prone to this issue, as the DDG API limits single IP concurrency, it is recommended to reduce concurrency and also use a proxy pool for requests.

Alternative go version deployment: free server serv00 deployment of resident services

Project source

1. Go to the serv00 panel and click on DNS zones and click on Add new zone to add your own customized domain name.
2. Go to cf to add A-record parsing
3. Click on Additional services and click on Run your own applications. Enabled
4. Open a port. I'm opening 5005.
5. Click WWW Websites Click Add website Domain to fill in your own domain name.

1. Click to download file Granting executable permissions to files

   chmod +x ddgchatgo-freebsd-amd64
   

2. existpublic_htmlCreate a startup script understart_ddgchatgo.sh

   #!/bin/bash
   # Define the process name and startup command
   PROCESS_NAME="ddgchatgo"
   START_COMMAND="PORT=5005 /home/yourServ00username/domains/example.com/public_html/ddgchatgo-freebsd-amd64 > /home/yourServ00user/domains/example .com/public_html/ddgchatgo.log 2>&1 &"
   # Check if the process is running
   if ! pgrep -f "$PROCESS_NAME" > /dev/null
   then
   echo "Process $PROCESS_NAME is not running, starting..."
   # Starting process
   eval "$START_COMMAND"
   if [ $? -eq 0 ]; then
   echo "Process $PROCESS_NAME started successfully."
   then echo "Process $PROCESS_NAME started successfully." else
   echo "Process $PROCESS_NAME failed."
   else echo "Process $PROCESS_NAME failed to start.
   echo "Process $PROCESS_NAME failed to start." fi
   echo "Process $PROCESS_NAME is already running."
   fi
   

3. Granting script execution privileges

   chmod +x start_api.sh
   

4. The first time you run cron, just run the following command

   nohup /home/your Serv00 username/domains/example.com/public_html/start_ddgchatgo.sh > /home/your Serv00 username/domains/example.com/public_html/start_ ddgchatgo.log 2>&1 &
   

   Change your Serv00 username to your own.
   example.com to your own domain name
   PORT If the open port is not 5005, you can change it.

One-Click Installation Scripts

#!/bin/bash
# Step 1: Enter the working directory
cd ~/domains/example.com/public_html/
# Step 2: Download the executable program
wget https://github.com/Shadownc/DDG-Chat-go/releases/download/v0.0.2/ddgchatgo-freebsd-amd64
# Step 3: Give executable permission to the extracted freechatgpt file
chmod +x ddgchatgo-freebsd-amd64
# Step 4: Create the start_ddgchatgo.sh script
cat < start_ddgchatgo.sh
#!/bin/bash
# Define the process name and start command
PROCESS_NAME="ddgchatgo"
START_COMMAND="PORT=5005 /home/your Serv00 username/domains/example.com/public_html/ddgchatgo-freebsd-amd64 > /home/your Serv00 username/domains/ example.com/public_html/ ddgchatgo-freebsd-amd64 > /home/your Serv00 username/domains/ example.com/public_html/ddgchatgo-freebsd-amd64 example.com/public_html/ddgchatgo.log 2>&1 &"
# Check if the process is running
if ! pgrep -f "$PROCESS_NAME" > /dev/null
then
echo "Process $PROCESS_NAME is not running, starting..."
# Starting process
eval "$START_COMMAND"
if [ $? -eq 0 ]; then
echo "Process $PROCESS_NAME started successfully."
then echo "Process $PROCESS_NAME started successfully." else
echo "Process $PROCESS_NAME failed to start."
echo "Process $PROCESS_NAME failed to start.
echo "Process $PROCESS_NAME failed to start." else
echo "Process $PROCESS_NAME is already running."
fi
EOL
# Step 5: Give start_ddgchatgo.sh executable privilege
chmod +x start_ddgchatgo.sh
# Step 6: Run start_ddgchatgo.sh script and output logs to start_ddgchatgo.log
nohup /home/your Serv00 username/domains/example.com/public_html/start_ddgchatgo.sh > /home/your Serv00 username/domains/example.com/public_html/start_ ddgchatgo.log 2>&1 &

In one-click scriptshttps://github.com/Shadownc/DDG-Chat-go/releases/download/v0.0.2/ddgchatgo-freebsd-amd64Replace withhttps://github.com/Shadownc/DDG-Chat-go/releases/download/v0.0.4/ddgchatgo-freebsd-amd64API KEY can be set.

Change your Serv00 username to your own.
example.com to your own domain name
PORT If the open port is not 5005, you can change it.

After deployment is complete, accessing the domain returns the following to indicate success:

{
"message": "Welcome to the API"
}

Access to newapi after building

new api/one api select custom channel
http://chat.xxl.serv00.net/v1/chat/completions

The layman's guide to getting started with AI

Help you learn how to utilize AI tools at a low cost and from a zero base.AI, like office software, is an essential skill for everyone. Mastering AI will give you an edge in your job search and half the effort in your future work and studies.

View Details>