Chatbot-GPT

Chatbot-GPT leverages OpenIM’s webhooks for seamless integration across messaging platforms, enhancing communication with intelligent, automated responses.
Live Demo

Contents

• Features
• Online Retrieval Architecture
• Deploy the OpenIM Service
• Deploy the Chatbot-GPT Service
  • Step 1: Download repository code
  • Step 2: Configure variables of .env
  • Step 3: Deploy Chatbot-GPT
    • Deploy Chatbot-GPT using Docker
    • Deploy Chatbot-GPT from source code
      • Set up the Python running environment
        • Create and activate a virtual environment
        • Install dependencies with pip
      • Upload Bot configuration
      • Start the service
• Register Chatbot-GPT
  • Register an account
  • Login and find the Bot

Features

• Integrated LLM Support: Incorporates both UnionLLM and LiteLLM, enabling simultaneous access to multiple large language models.
• Quick Setup: Deploys production-level conversational service robots in just five minutes.
• Real-World Application: Optimally designed for real-time customer service systems to tackle genuine business challenges.
• Attractive UI: Offers a customizable and visually compelling user interface.

Online Retrieval Architecture

Deploy the OpenIM Service

Step 1: Download repository code

Clone the repository:

git clone https://github.com/openimsdk/openim-docker && cd openim-docker

Step 2: Configure variables of .env

Modify the .env file to set up the external IP address for the MinIO and Grafana services.

# Set the external access address for MinIO service (external IP or domain)
MINIO_EXTERNAL_ADDRESS="http://external_ip:10005"
# Set the external access address for Grafana service (external IP or domain)
GRAFANA_URL="http://external_ip:13000/"

Step 3: Configure webhooks for Chatbot-GPT

Modify openim-server environment variables in docker-compose.yaml.

    openim-server:
      ...
      environment:
        - IMENV_WEBHOOKS_URL=http://127.0.0.1:9000/open_chatbot
        - IMENV_WEBHOOKS_AFTERSENDSINGLEMSG_ENABLE=true
        - IMENV_WEBHOOKS_AFTERSENDGROUPMSG_ENABLE=true
        ...

Step 4: Start the OpenIM Service

docker compose up -d

More details about OpenIM deploy, please refer to OpenIM Docs

Deploy the Chatbot-GPT Service

Step 1: Download repository code

Clone the repository:

git clone https://github.com/gpt-open/chatbot-gpt.git && cd chatbot-gpt

Step 2: Configure variables of .env

Before starting the Chatbot-GPT service, you need to modify the related configurations for the program to initialize correctly.

cp env_example .env

The variables in .env

GET_TOKEN_URL="http://127.0.0.1:10009/account/login"
ADD_BOT_URL="http://127.0.0.1:10009/user/import/json"
UPDATE_BOT_URL="http://127.0.0.1:10008/user/update"
SEND_MSG_URL="http://127.0.0.1:10002/msg/send_msg"
TOKEN_EXPIRE_DAYS=30
MAX_HISTORY_SESSION_LENGTH=3

Note

Please modify the GET_TOKEN_URL, ADD_BOT_URL, UPDATE_BOT_URL, SEND_MSG_URL, and TOKEN_EXPIRE_DAYS in conjunction with the OpenIM Server configuration.

Step 3: Deploy Chatbot-GPT

Deploy Chatbot-GPT using Docker

docker-compose up --build

Deploy Chatbot-GPT from source code

Note

Please use Python version 3.10.x or above.

Set up the Python running environment

It is recommended to install Python-related dependencies in a Python virtual environment to avoid affecting dependencies of other projects.

Create and activate a virtual environment

If you have not yet created a virtual environment, you can create one with the following command:

python3 -m venv myenv

After creation, activate the virtual environment:

source myenv/bin/activate

Install dependencies with pip

Once the virtual environment is activated, you can use pip to install the required dependencies.

pip install -r requirements.txt

Upload Bot configuration

Note

Chatbot-GPT can integrate any number of LLM Bots.

Please refer to bot_config.yaml and modify this configuration file according to your actual needs.

userID in bot_config.yaml must be unique.

robots:
  OpenAI:
    - apiKey: "xxxx"
      model: "gpt-3.5-turbo"
      userID: "openai_robot1"
      nickname: "ChatGPT 3.5"
      faceURL: "https://openai.xiniushu.com/img/logo.png"
    - apiKey: "xxxx"
      model: "gpt-4o"
      userID: "openai_robot2"
      nickname: "ChatGPT 4o"
      faceURL: "https://openai.xiniushu.com/img/logo.png"

  ZhipuAI:
    - apiKey: "xxxx"
      model: "glm-3-turbo"
      userID: "zhipuai_robot1"
      nickname: "GLM-3"
      faceURL: "https://pp.myapp.com/ma_icon/0/icon_54321899_1717467379/256"
    - apiKey: "xxxx"
      model: "glm-4"
      userID: "zhipuai_robot2"
      nickname: "GLM-4"
      faceURL: "https://pp.myapp.com/ma_icon/0/icon_54321899_1717467379/256"

  Moonshot:
    - apiKey: "xxxx"
      model: "moonshot-v1-32k"
      userID: "moonshot_robot1"
      nickname: "Moonshot-chat"
      faceURL: "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcSnsfyv3cjhrqvNb-Mos6InwLI4Pnjdxiw8CVtxA-tRSlRWwl-nba84&usqp=CAE&s"

  Deepseek:
    - apiKey: "xxxx"
      model: "deepseek/deepseek-chat"
      userID: "deepseek_robot1"
      nickname: "Deepseek-chat"
      faceURL: "https://avatars.githubusercontent.com/u/148330874?s=200&v=4"

python3 upload_bot_config.py

Start the service

If you have completed the steps above, you can try to start the Chatbot-GPT service by executing the following command.

• Start single process:

python3 chatbot_app.py

• Start multiple processes:

sh start.sh

Note

• The service port for Chatbot-GPT is 9000. During the first test, please try not to change the port so that you can quickly experience the entire product process.
• We recommend starting the Chatbot-GPT service using start.sh in multi-process mode for a smoother user experience.

Register Chatbot-GPT

Register an account

Please visit the link http://127.0.0.1:9000/#/login and register using your mobile number. The demo here does not require SMS verification code validation.

Login and find the Bot

Log in to Chatbot-GPT using your username and password, then you can add friends or join groups.