OpenAi-HK

简体中文

English

简体中文

English

Appearance

快速接入 ​

  • 目前 OpenAI-HK 提供 GPT、Dall-e-3、gpt-4o o1 o3 gpt-image-1 等 openai 中转接口服务，需要支付费用后才可调用

  • 同时 OpenAI-HK 还提供 Suno、Flux、Midjourney、Google Veo3、Riffusion 等涵盖 AI Music、AI Video、AI Image、AI Chat 服务

  • 官方参考 需要科学上网

调用方式 ​

  1. 先通过 openai-hk注册一个帐号；
  2. 登录后，点击：【获取 key】即可获得 api-key;
  3. 复制 KEY  返回的 hk-开头的 api_key，将原请求头中的 api_key  替换成我们专属的  api_key cover 例如：

原来: sk-sdiL2SMN4D7GBqFhBsYdT2345kFJhwEHGXbU5RzCM8dS3lrn

现在: hk-wsvj0oyeb0srl8b76xgzolc678996mhwlk3p3y7p3qo9wjdl

OpenAI  的  api_key  通常以  sk-  作为开头，但我们平台的  api_key  以  hk-  开头   后面是  48  位随机字符串

原请求地址  api.openai.com 的后面加上-hk，即  api.openai-hk.com

例如：

原来: https://api.openai.com/v1/chat/completions

现在: https://api.openai-hk.com/v1/chat/completions

🔥 对话接口实例 ​

CURL  实例 ​

shell
curl --request POST \  --url https://api.openai-hk.com/v1/chat/completions \  --header 'Authorization: Bearer hk-替换为你的key' \  -H "Content-Type: application/json" \  --data '{  "max_tokens": 1200,  "model": "gpt-3.5-turbo",  "temperature": 0.8,  "top_p": 1,  "presence_penalty": 1,  "messages": [         {  "role": "system",  "content": "You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible."         },         {  "role": "user",  "content": "你是chatGPT多少？"         }     ] }'
curl --request POST \  --url https://api.openai-hk.com/v1/chat/completions \  --header 'Authorization: Bearer hk-替换为你的key' \  -H "Content-Type: application/json" \  --data '{  "max_tokens": 1200,  "model": "gpt-3.5-turbo",  "temperature": 0.8,  "top_p": 1,  "presence_penalty": 1,  "messages": [         {  "role": "system",  "content": "You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible."         },         {  "role": "user",  "content": "你是chatGPT多少？"         }     ] }'

如需调用 4.0，可将上面 modle 中 gpt-3.5-turbo 改为 gpt-4 或 gpt-4-0613 即可。

PHP 实例 ​

php
 $ch = curl_init();  curl_setopt($ch, CURLOPT_URL, 'https://api.openai-hk.com/v1/chat/completions'); curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1); curl_setopt($ch, CURLOPT_POST, 1);  $headers = array(); $headers[] = 'Content-Type: application/json'; $headers[] = 'Authorization: Bearer hk-替换为你的key'; curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);  $data = array(  'max_tokens' => 1200,  'model' => 'gpt-3.5-turbo',  'temperature' => 0.8,  'top_p' => 1,  'presence_penalty' => 1,  'messages' => array(  array(  'role' => 'system',  'content' => 'You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible.'  ),  array(  'role' => 'user',  'content' => '你是chatGPT多少？'  )  ) );  $data_string = json_encode($data);  curl_setopt($ch, CURLOPT_POSTFIELDS, $data_string);  $result = curl_exec($ch);  if (curl_errno($ch)) {  echo 'Error:' . curl_error($ch); }  curl_close($ch);  echo $result;
 $ch = curl_init();  curl_setopt($ch, CURLOPT_URL, 'https://api.openai-hk.com/v1/chat/completions'); curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1); curl_setopt($ch, CURLOPT_POST, 1);  $headers = array(); $headers[] = 'Content-Type: application/json'; $headers[] = 'Authorization: Bearer hk-替换为你的key'; curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);  $data = array(  'max_tokens' => 1200,  'model' => 'gpt-3.5-turbo',  'temperature' => 0.8,  'top_p' => 1,  'presence_penalty' => 1,  'messages' => array(  array(  'role' => 'system',  'content' => 'You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible.'  ),  array(  'role' => 'user',  'content' => '你是chatGPT多少？'  )  ) );  $data_string = json_encode($data);  curl_setopt($ch, CURLOPT_POSTFIELDS, $data_string);  $result = curl_exec($ch);  if (curl_errno($ch)) {  echo 'Error:' . curl_error($ch); }  curl_close($ch);  echo $result;

如需调用 4.0，可将上面 modle 中 gpt-3.5-turbo 改为 gpt-4 或 gpt-4-0613 即可。

Python 实例 ​

Python
import requests import json  url = "https://api.openai-hk.com/v1/chat/completions"  headers = {  "Content-Type": "application/json",  "Authorization": "Bearer hk-替换为你的key" }  data = {  "max_tokens": 1200,  "model": "gpt-3.5-turbo",  "temperature": 0.8,  "top_p": 1,  "presence_penalty": 1,  "messages": [  {  "role": "system",  "content": "You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible."  },  {  "role": "user",  "content": "你是chatGPT多少？"  }  ] }  response = requests.post(url, headers=headers, data=json.dumps(data).encode('utf-8') ) result = response.content.decode("utf-8")  print(result)
import requests import json  url = "https://api.openai-hk.com/v1/chat/completions"  headers = {  "Content-Type": "application/json",  "Authorization": "Bearer hk-替换为你的key" }  data = {  "max_tokens": 1200,  "model": "gpt-3.5-turbo",  "temperature": 0.8,  "top_p": 1,  "presence_penalty": 1,  "messages": [  {  "role": "system",  "content": "You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible."  },  {  "role": "user",  "content": "你是chatGPT多少？"  }  ] }  response = requests.post(url, headers=headers, data=json.dumps(data).encode('utf-8') ) result = response.content.decode("utf-8")  print(result)

如需调用 4.0，可将上面 modle 中 gpt-3.5-turbo 改为 gpt-4 或 gpt-4-0613 即可。

javascript 实例 ​

js
const axios = require("axios");  const url = "https://api.openai-hk.com/v1/chat/completions"; const headers = {  "Content-Type": "application/json",  Authorization: "Bearer hk-替换为你的key", };  const data = {  max_tokens: 1200,  model: "gpt-3.5-turbo",  temperature: 0.8,  top_p: 1,  presence_penalty: 1,  messages: [  {  role: "system",  content:  "You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible.",  },  {  role: "user",  content: "你是chatGPT多少？",  },  ], };  axios  .post(url, data, { headers })  .then((response) => {  const result = response.data;  console.log(result);  })  .catch((error) => {  console.error(error);  });
const axios = require("axios");  const url = "https://api.openai-hk.com/v1/chat/completions"; const headers = {  "Content-Type": "application/json",  Authorization: "Bearer hk-替换为你的key", };  const data = {  max_tokens: 1200,  model: "gpt-3.5-turbo",  temperature: 0.8,  top_p: 1,  presence_penalty: 1,  messages: [  {  role: "system",  content:  "You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible.",  },  {  role: "user",  content: "你是chatGPT多少？",  },  ], };  axios  .post(url, data, { headers })  .then((response) => {  const result = response.data;  console.log(result);  })  .catch((error) => {  console.error(error);  });

如需调用 4.0，可将上面 modle 中 gpt-3.5-turbo 改为 gpt-4 或 gpt-4-0613 即可。

typescript 实例 ​

typescript
import axios from "axios";  const url = "https://api.openai-hk.com/v1/chat/completions"; const headers = {  "Content-Type": "application/json",  Authorization: "Bearer hk-替换为你的key", };  const data = {  max_tokens: 1200,  model: "gpt-3.5-turbo",  temperature: 0.8,  top_p: 1,  presence_penalty: 1,  messages: [  {  role: "system",  content:  "You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible.",  },  {  role: "user",  content: "你是chatGPT多少？",  },  ], };  axios  .post(url, data, { headers })  .then((response) => {  const result = response.data;  console.log(result);  })  .catch((error) => {  console.error(error);  });
import axios from "axios";  const url = "https://api.openai-hk.com/v1/chat/completions"; const headers = {  "Content-Type": "application/json",  Authorization: "Bearer hk-替换为你的key", };  const data = {  max_tokens: 1200,  model: "gpt-3.5-turbo",  temperature: 0.8,  top_p: 1,  presence_penalty: 1,  messages: [  {  role: "system",  content:  "You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible.",  },  {  role: "user",  content: "你是chatGPT多少？",  },  ], };  axios  .post(url, data, { headers })  .then((response) => {  const result = response.data;  console.log(result);  })  .catch((error) => {  console.error(error);  });

如需调用 4.0，可将上面 modle 中 gpt-3.5-turbo 改为 gpt-4 或 gpt-4-0613 即可。

java 实例 ​

java
import okhttp3.*;  import java.io.IOException;  public class OpenAIChat {   public static void main(String[] args) throws IOException {  String url = "https://api.openai-hk.com/v1/chat/completions";   OkHttpClient client = new OkHttpClient();   MediaType mediaType = MediaType.parse("application/json");   String json = "{\n" +  " \"max_tokens\": 1200,\n" +  " \"model\": \"gpt-3.5-turbo\",\n" +  " \"temperature\": 0.8,\n" +  " \"top_p\": 1,\n" +  " \"presence_penalty\": 1,\n" +  " \"messages\": [\n" +  " {\n" +  " \"role\": \"system\",\n" +  " \"content\": \"You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible.\"\n" +  " },\n" +  " {\n" +  " \"role\": \"user\",\n" +  " \"content\": \"你是chatGPT多少？\"\n" +  " }\n" +  " ]\n" +  "}";   RequestBody body = RequestBody.create(mediaType, json);  Request request = new Request.Builder()  .url(url)  .post(body)  .addHeader("Content-Type", "application/json")  .addHeader("Authorization", "Bearer hk-替换为你的key")  .build();   Call call = client.newCall(request);  Response response = call.execute();  String result = response.body().string();  System.out.println(result);  }  }
import okhttp3.*;  import java.io.IOException;  public class OpenAIChat {   public static void main(String[] args) throws IOException {  String url = "https://api.openai-hk.com/v1/chat/completions";   OkHttpClient client = new OkHttpClient();   MediaType mediaType = MediaType.parse("application/json");   String json = "{\n" +  " \"max_tokens\": 1200,\n" +  " \"model\": \"gpt-3.5-turbo\",\n" +  " \"temperature\": 0.8,\n" +  " \"top_p\": 1,\n" +  " \"presence_penalty\": 1,\n" +  " \"messages\": [\n" +  " {\n" +  " \"role\": \"system\",\n" +  " \"content\": \"You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible.\"\n" +  " },\n" +  " {\n" +  " \"role\": \"user\",\n" +  " \"content\": \"你是chatGPT多少？\"\n" +  " }\n" +  " ]\n" +  "}";   RequestBody body = RequestBody.create(mediaType, json);  Request request = new Request.Builder()  .url(url)  .post(body)  .addHeader("Content-Type", "application/json")  .addHeader("Authorization", "Bearer hk-替换为你的key")  .build();   Call call = client.newCall(request);  Response response = call.execute();  String result = response.body().string();  System.out.println(result);  }  }

如需调用 4.0，可将上面 modle 中 gpt-3.5-turbo 改为 gpt-4 或 gpt-4-0613 即可。

go 实例 ​

go
package main  import ( "bytes" "encoding/json" "fmt" "net/http" )  func main() { url := "https://api.openai-hk.com/v1/chat/completions" apiKey := "hk-替换为你的key"  payload := map[string]interface{}{ "max_tokens": 1200, "model": "gpt-3.5-turbo", "temperature": 0.8, "top_p": 1, "presence_penalty": 1, "messages": []map[string]string{ { "role": "system", "content": "You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible.", }, { "role": "user", "content": "你是chatGPT多少？", }, }, }  jsonPayload, err := json.Marshal(payload) if err != nil { fmt.Println("Error encoding JSON payload:", err) return }  req, err := http.NewRequest("POST", url, bytes.NewBuffer(jsonPayload)) if err != nil { fmt.Println("Error creating HTTP request:", err) return }  req.Header.Set("Authorization", "Bearer "+apiKey) req.Header.Set("Content-Type", "application/json")  client := &http.Client{} resp, err := client.Do(req) if err != nil { fmt.Println("Error making API request:", err) return } defer resp.Body.Close()  // 处理响应 // 请根据实际需求解析和处理响应数据 fmt.Println("Response HTTP Status:", resp.StatusCode) }
package main  import ( "bytes" "encoding/json" "fmt" "net/http" )  func main() { url := "https://api.openai-hk.com/v1/chat/completions" apiKey := "hk-替换为你的key"  payload := map[string]interface{}{ "max_tokens": 1200, "model": "gpt-3.5-turbo", "temperature": 0.8, "top_p": 1, "presence_penalty": 1, "messages": []map[string]string{ { "role": "system", "content": "You are ChatGPT, a large language model trained by OpenAI. Answer as concisely as possible.", }, { "role": "user", "content": "你是chatGPT多少？", }, }, }  jsonPayload, err := json.Marshal(payload) if err != nil { fmt.Println("Error encoding JSON payload:", err) return }  req, err := http.NewRequest("POST", url, bytes.NewBuffer(jsonPayload)) if err != nil { fmt.Println("Error creating HTTP request:", err) return }  req.Header.Set("Authorization", "Bearer "+apiKey) req.Header.Set("Content-Type", "application/json")  client := &http.Client{} resp, err := client.Do(req) if err != nil { fmt.Println("Error making API request:", err) return } defer resp.Body.Close()  // 处理响应 // 请根据实际需求解析和处理响应数据 fmt.Println("Response HTTP Status:", resp.StatusCode) }

如需调用 4.0，可将上面 modle 中 gpt-3.5-turbo 改为 gpt-4 或 gpt-4-0613 即可。

🚀js 实现 sse 打字效果 ​

注意

可 node.js 后端运行；也支持浏览器模式；需要注意的是如果是前端 注意保护好你的 key。

如何保护好 key？可以跟 nginx 配合 将 header 部分的 'Authorization': 'Bearer hk-你的key' 放到 nginx 当中

js
//记得引入 `axios` const chatGPT = (msg, opt) => {  let content = "";  const dataPar = (data) => {  let rz = {};  let dz = data.split("data:", 2);  const str = dz[1].trim();  if (str == "[DONE]") rz = { finish: true, text: "" };  else {  rz = JSON.parse(str);  rz.text = rz.choices[0].delta.content;  }  return rz;  };  const dd = (data) => {  let arr = data.trim().split("\n\n");  let rz = { text: "", arr: [] };  const atext = arr.map((v) => {  const aa = dataPar(v);  return aa.text;  });  rz.arr = atext;  rz.text = atext.join("");  if (opt.onMessage) opt.onMessage(rz);  return rz;  };  return new Promise((resolve, reject) => {  axios({  method: "post",  url: "https://api.openai-hk.com/v1/chat/completions",  data: {  max_tokens: 1200,  model: "gpt-3.5-turbo", //模型替换 如需调用4.0，改为gpt-4或gpt-4-0613即可  temperature: 0.8,  top_p: 1,  presence_penalty: 1,  messages: [  {  role: "system",  content: opt.system ?? "You are ChatGPT",  },  {  role: "user",  content: msg,  },  ],  stream: true, //数据流方式输出  },  headers: {  "Content-Type": "application/json",  Authorization: "Bearer hk-你的key",  },  onDownloadProgress: (e) => dd(e.target.responseText),  })  .then((d) => resolve(dd(d.data)))  .catch((e) => reject(e));  }); };  //调用 chatGPT("你是谁？", {  //system:'', //角色定义  onMessage: (d) => console.log("过程性结果：", d.text), })  .then((d) => console.log("✅最终结果：", d))  .catch((e) => console.log("❎错误：", e));
//记得引入 `axios` const chatGPT = (msg, opt) => {  let content = "";  const dataPar = (data) => {  let rz = {};  let dz = data.split("data:", 2);  const str = dz[1].trim();  if (str == "[DONE]") rz = { finish: true, text: "" };  else {  rz = JSON.parse(str);  rz.text = rz.choices[0].delta.content;  }  return rz;  };  const dd = (data) => {  let arr = data.trim().split("\n\n");  let rz = { text: "", arr: [] };  const atext = arr.map((v) => {  const aa = dataPar(v);  return aa.text;  });  rz.arr = atext;  rz.text = atext.join("");  if (opt.onMessage) opt.onMessage(rz);  return rz;  };  return new Promise((resolve, reject) => {  axios({  method: "post",  url: "https://api.openai-hk.com/v1/chat/completions",  data: {  max_tokens: 1200,  model: "gpt-3.5-turbo", //模型替换 如需调用4.0，改为gpt-4或gpt-4-0613即可  temperature: 0.8,  top_p: 1,  presence_penalty: 1,  messages: [  {  role: "system",  content: opt.system ?? "You are ChatGPT",  },  {  role: "user",  content: msg,  },  ],  stream: true, //数据流方式输出  },  headers: {  "Content-Type": "application/json",  Authorization: "Bearer hk-你的key",  },  onDownloadProgress: (e) => dd(e.target.responseText),  })  .then((d) => resolve(dd(d.data)))  .catch((e) => reject(e));  }); };  //调用 chatGPT("你是谁？", {  //system:'', //角色定义  onMessage: (d) => console.log("过程性结果：", d.text), })  .then((d) => console.log("✅最终结果：", d))  .catch((e) => console.log("❎错误：", e));

😁embeddings 接口 ​

请求地址： POST https://api.openai-hk.com/v1/embeddings

node.js 请求实例

js
const fetch = require("node-fetch");  fetch("https://api.openai-hk.com/v1/embeddings", {  method: "POST",  headers: {  Authorization: "Bearer hk-替换为你的key",  "Content-Type": "application/json",  },  body: JSON.stringify({  input: "一起来使用ChatGPT",  model: "text-embedding-ada-002",  }), });
const fetch = require("node-fetch");  fetch("https://api.openai-hk.com/v1/embeddings", {  method: "POST",  headers: {  Authorization: "Bearer hk-替换为你的key",  "Content-Type": "application/json",  },  body: JSON.stringify({  input: "一起来使用ChatGPT",  model: "text-embedding-ada-002",  }), });

😁moderations 接口 ​

请求地址： POST https://api.openai-hk.com/v1/moderations

node.js 请求实例

js
const fetch = require("node-fetch");  fetch("https://api.openai-hk.com/v1/moderations", {  method: "POST",  headers: {  Authorization: "Bearer hk-替换为你的key",  "Content-Type": "application/json",  },  body: JSON.stringify({ input: "有人砍我" }), });
const fetch = require("node-fetch");  fetch("https://api.openai-hk.com/v1/moderations", {  method: "POST",  headers: {  Authorization: "Bearer hk-替换为你的key",  "Content-Type": "application/json",  },  body: JSON.stringify({ input: "有人砍我" }), });

😁 各种应用 ​

现实中有好多应用，可以选择一种你喜欢的

chatgpt-web ​

docker 启动 默认模型是 gpt-3.5

shell
docker run --name chatgpt-web -d -p 6011:3002 \ --env OPENAI_API_KEY=hk-替换为你的key \ --env TIMEOUT_MS=600000 --env OPENAI_MAX_TOKEN=1000 \ --env OPENAI_API_BASE_URL=https://api.openai-hk.com chenzhaoyu94/chatgpt-web
docker run --name chatgpt-web -d -p 6011:3002 \ --env OPENAI_API_KEY=hk-替换为你的key \ --env TIMEOUT_MS=600000 --env OPENAI_MAX_TOKEN=1000 \ --env OPENAI_API_BASE_URL=https://api.openai-hk.com chenzhaoyu94/chatgpt-web

然后访问 http://127.0.0.1:6011

chatgpt-web gpt-4 ​

默认模型是 gpt-3.5 如何起一个默认模型 gpt-4.0 呢? 使用环境变量 OPENAI_API_MODEL

shell
docker run --name chatgpt-web -d -p 6040:3002 \ --env OPENAI_API_KEY=hk-替换为你的key \ --env TIMEOUT_MS=600000 --env OPENAI_MAX_TOKEN=1000 \ --env OPENAI_API_MODEL=gpt-4-0613 \ --env OPENAI_API_BASE_URL=https://api.openai-hk.com chenzhaoyu94/chatgpt-web
docker run --name chatgpt-web -d -p 6040:3002 \ --env OPENAI_API_KEY=hk-替换为你的key \ --env TIMEOUT_MS=600000 --env OPENAI_MAX_TOKEN=1000 \ --env OPENAI_API_MODEL=gpt-4-0613 \ --env OPENAI_API_BASE_URL=https://api.openai-hk.com chenzhaoyu94/chatgpt-web

然后访问 http://127.0.0.1:6040

可选模型有哪些 ​

模型说明
gpt-3.5-turbo-06133.5 的 4k 上下文支持到 4k
gpt-3.5-turbo-16k-06133.5 的 16k 能够支持更长的上下文 价格比普通 3.5 贵些
gpt-3.5-turbo-11063.5 的 16k 价格与 3.5 4k 一样
gpt-48k 的 4.0 上下文支持到 8k
gpt-4-06134.0 的 0613 版本
gpt-4-1106-preview128k 的 4.0 价格是普通 gpt-4 的一半
gpt-4-vision-previewtoken 价格跟 gpt-4-1106-preview 一样，支持 4k, 有额外的图片费用
dall-e-3openai 的画图
Midjourney非 openai 的画图产品

看效果 cover

chatgpt-next-web ​

shell
docker run --name chatgpt-next-web -d -p 6013:3000 \  -e OPENAI_API_KEY="hk-替换为你的key" \  -e BASE_URL=https://api.openai-hk.com yidadaa/chatgpt-next-web
docker run --name chatgpt-next-web -d -p 6013:3000 \  -e OPENAI_API_KEY="hk-替换为你的key" \  -e BASE_URL=https://api.openai-hk.com yidadaa/chatgpt-next-web

然后访问 http://127.0.0.1:6013

看效果

cover

其他 ​

如果还有其应用 请联系我们客服协助

🚀 直接使用 ​

1、chatgpt web 多模型自由切换 ​

注意 1.访问 https://chat.ddaiai.com/ (如果被墙，请换用将二级域名 chat更换为 hi)


2.如果发现被墙了 可以自己换地址 https://hello.ddaiai.com 把 hello 换其他的 如 https://202312.ddaiai.com 都能访问

设置

然后在下图相应的地方设置 OpenAi接口地址: https://api.openai-hk.com OpenAi API KEY: hk-你的apiKey  如果想画图 把midjourney的资料也完善 midjourney 接口地址: https://api.openai-hk.com midjourney Api Secret: hk-你的apiKey
然后在下图相应的地方设置 OpenAi接口地址: https://api.openai-hk.com OpenAi API KEY: hk-你的apiKey  如果想画图 把midjourney的资料也完善 midjourney 接口地址: https://api.openai-hk.com midjourney Api Secret: hk-你的apiKey

cover效果

输入框 提问对话 cover

模型切换 支持自定义模型，多个可以用逗号隔开 cover

2、next-web 多模型自由切换 ​

注意 1.访问 https://web.ccaiai.com/ (需要科学)


2.如果发现被墙了 可以自己换地址 https://suibian.ccaiai.com 把 suibian 换其他的 如 https://abc.ccaiai.com 都能访问

然后在下图相应的地方设置 接口地址：https://api.openai-hk.com API KEY: hk-你的apiKey 模型：选择 GPT-4
然后在下图相应的地方设置 接口地址：https://api.openai-hk.com API KEY: hk-你的apiKey 模型：选择 GPT-4

cover

输入框 提问对话

cover