Docker Project: ChatGPT API Platform

This document is also supported for the following languages

English | 简体中文 | 繁體中文 | Русский

Self Built

Copy the following code and modify the configuration yourself

version: '3'

# GPL protocol open source, no commercial use, no modification and closure after modification, no profit!!!

services:
  web:
    image: sengedev/chatgpt:latest   # Mirror address
    ports:
      - "5000:5000" # Port number, if the port conflicts, please modify it
    restart: always # Restart strategy
    environment:
      API_KEY: YourOpenAIApiKey # Please replace it with your OpenAI API Key, go to https://platform.openai.com/account/api-keys to get it
      HOUR_LIMIT: 50  # Limit the number of calls per hour, if you don't want to limit it, don't set this variable
      MINUTE_LIMIT: 3 # Limit the number of calls per minute, if you don't want to limit it, don't set this variable
      SECOND_LIMIT: 1 # Limit the number of calls per second, if you don't want to limit it, don't set this variable
      ROUTE: api  # Route, if you don't want to set it, don't set this environment variable
      LANG: zh_CN # Language, currently supports Simplified Chinese, Traditional Chinese, English, and Russian

Field meaning

Field	Meaning
API_KEY	API key, after configuration, the call can be completed without the request header, it is not recommended to set it, unless you have enabled the IP address whitelist
HOUR_LIMIT	The number of calls per hour is limited. If it is set to 0, there is no limit
MINUTE_LIMIT	The number of calls per minute is limited. If it is set to 0, there is no limit
SECOND_LIMIT	The number of calls per second is limited. If it is set to 0, there is no limit
ROUTE	For example, if your website is https://chatgpt.example.com and the route is api, the request link is https://chatgpt.example.com/api

Instance Introduction

Tutorial

Here is an example API address, please replace it with your own IP/domain.

API link: https://chatgpt.example.com/

Get ChatGPT answer

Request method: POST, route: /api

Request Parameters

Data	Description	Required
sys_content	System configuration, some restrictions on ChatGPT	No
user_content	Content sent to ChatGPT	Yes
model	Model, default to text-davinci-003 model	No
api_key	API key, must be configured, unless you add an API key in Docker-comppose	No
max_tokens	The maximum number of tokens generated, default is 100, maximum is 3500	No
continuous	Continuous dialogue parameters, can be iterated to achieve context dialogue operations	No

Response

• Success

{
    "code": 200,
    "msg": "success",
    "data": {
        "response": "Content answered by ChatGPT"
    }
}

• Failure

{
    "code": "4xx",
    "msg": "failed",
    "data": {
        "response": "Reason for request failure"
    }
}

• Common response values

code	Description
200	Success
400	Request parameter error (API not filled in or filled in incorrectly, model usage error, prompt missing or prompt empty)
401	Unauthorized
403	Forbidden
404	Request path does not exist
500	Internal server error
429	Request too frequent

Default call limit (per IP)

The API is completely open for use without application, but you need to apply for an API Key yourself.

Time period	Frequency
Day	Unlimited
Hour	50 times
Minute	3 times
Second	1 time

Example Code

Replace https://chatgpt.example.com/api with your server IP/domain address

Python (Continuous Dialogue)

import requests
import json


token = {
    "api_key": "Your API Key",  # Replace with your own API Key
    "url": "https://chatgpt.example.com/api"    # Replace with your own instance
}


# ChatGPT API Demo
url = token["url"]
sys_prompt = input("sys> ")

continuous_dialogue = [{"role": "system", "content": sys_prompt}]
while True:
    user_input = input("user> ")
    if user_input.lower() in ['exit', 'quit']:
        break
    data = {
        "system_content": sys_prompt,
        "user_content": user_input,
        "model": "gpt-3.5-turbo",
        "api_key": token["api_key"],
        "continuous": [],
        "max_tokens": 3000
    }
    response = requests.post(url, json=data)
    if response.status_code != 200:
        print(f"Request failed, status code: {response.status_code}, server response: {response.text}")
        continue
    else:
        result = json.loads(response.text)
    print(f"ChatGPT> {result['current_response']}")
    # Continuous dialogue iteration
    continuous_dialogue.append({"role": "user", "content": user_input})
    continuous_dialogue.append({"role": "assistant", "content": result['current_response']})

print("Program has exited")

Python (Non-Continuous Dialogue)

import sys
import requests
import json


token = {
    "api_key": "Your API Key",  # Replace with your own API Key
    "url": "https://chatgpt.example.com/api"    # Replace with your own instance
}


# ChatGPT API Demo
url = token["url"]
sys_prompt = input("sys> ")

user_input = input("user> ")
data = {
    "system_content": sys_prompt,
    "user_content": user_input,
    "model": "gpt-3.5-turbo",
    "api_key": token["api_key"],
    "continuous": [],
    "max_tokens": 3000
}
response = requests.post(url, json=data)
if response.status_code != 200:
    print(f"Request failed, status code: {response.status_code}, server response: {response.text}")
    sys.exit(1)
else:
    result = json.loads(response.text)
print(f"ChatGPT> {result['current_response']}")