claude3国内API接口对接

众所周知，由于地理位置原因，Claude3不对国内开放，而国内的镜像网站使用又贵的离谱！

因此，团队萌生了一个想法：为什么不创建一个一站式的平台，让用户能够通过单一的接口与多个模型交流呢？这样，用户就可以轻松地比较不同模型的表现，并根据需要选择最合适的一个。于是诞生了这个ChatGPT，Claude3聚合API平台

API对接网站：api.atalk-ai.com/

话不多说，先上截图，可以看到不管是ChatGPT还是Claude3的模型它都支持

通过一个接口即可对接国际主流AI模型，兼容性这边已经帮大家处理好了，无脑对接即可 API文档地址：api.atalk-ai.com/api#/operat...

对接方式

Python

python 复制代码

import http.client

conn = http.client.HTTPSConnection("api.atalk-ai.com")

payload = "{\n \"messages\": [\n {\n \"role\": \"system\",\n \"content\": \"You are a helpful assistant.\"\n },\n {\n \"role\": \"assistant\",\n \"content\": \"can i help you ?\"\n },\n {\n \"role\": \"user\",\n \"content\": \"Hello!\"\n }\n ],\n \"model\": \"gpt-3.5-turbo\",\n \"max_tokens\": 1000,\n \"stream\": true,\n \"temperature\": 0.2\n}"

headers = {

'Authorization': "",

'Content-Type': "application/json"

}

conn.request("POST", "/gpt/completions", payload, headers)

res = conn.getresponse()

data = res.read()

print(data.decode("utf-8"))

java

java 复制代码

AsyncHttpClient client = new DefaultAsyncHttpClient();

client.prepare("POST", "https://api.atalk-ai.com/gpt/completions")

.setHeader("Authorization", "")

.setHeader("Content-Type", "application/json")

.setBody("{\n \"messages\": [\n {\n \"role\": \"system\",\n \"content\": \"You are a helpful assistant.\"\n },\n {\n \"role\": \"assistant\",\n \"content\": \"can i help you ?\"\n },\n {\n \"role\": \"user\",\n \"content\": \"Hello!\"\n }\n ],\n \"model\": \"gpt-3.5-turbo\",\n \"max_tokens\": 1000,\n \"stream\": true,\n \"temperature\": 0.2\n}")

.execute()

.toCompletableFuture()

.thenAccept(System.out::println)

.join();

client.close();

PHP

php 复制代码

<?php

$client = new \GuzzleHttp\Client();

$response = $client->request('POST', 'https://api.atalk-ai.com/gpt/completions', [

'body' => '{

"messages": [

{

"role": "system",

"content": "You are a helpful assistant."

},

{

"role": "assistant",

"content": "can i help you ?"

},

{

"role": "user",

"content": "Hello!"

}

],

"model": "gpt-3.5-turbo",

"max_tokens": 1000,

"stream": true,

"temperature": 0.2

}',

'headers' => [

'Authorization' => '',

'Content-Type' => 'application/json',

],

]);

echo $response->getBody();

海鲸AI-API聚合平台是我们对AI技术无限探索的一次尝试。它不仅简化了与多个AI模型的交互过程，也为用户提供了一个高效、便捷的解决方案。我相信，随着AI技术的不断进步，海鲸AI将成为您实现创意和解决问题的得力助手。