Retrying langchain.chat_models.openai.ChatOpenAI.completion_with_retry.._completion_with_retry in 1.0 seconds as it raised APIError: Invalid response object from API: 'An error occurred with this application.\n\nROUTER_EXTERNAL_TARGET_ERROR\n' (HTTP response code was 502).

请问这个和通过vercel转发有关系么

用apifox 调用会出现
An error occurred with this application.

ROUTER_EXTERNAL_TARGET_ERROR

This content violates Vercel's Terms of Service, and/or rules on Fair Use.

你们有遇见过吗，。在部署的时候

您好，我在按您的教程配置完成后仍然出现了如下问题，想咨询一下解决方案：
openai.error.InvalidRequestError: Invalid URL (POST/v1/chat/completions/chat/completions)

代码如下：
import openai

openai.api_key = "****"
host = '*****'
url = f'https://{host}/v1/chat/completions'
openai.api_base = url
messages = [
{"role": "system", "content": "You are a news creator"},
{"role": "user",
"content": f"Write a news article with the title 你好 that spans 3-4 paragraphs and comprehensively covers pertinent and vital information on the topic. Your piece should be both informative and engaging, capturing readers' attention and keeping them informed throughout."},
]

调用 GPT-3.5-turbo API

completion = openai.ChatCompletion.create(
model="gpt-3.5-turbo",
messages=messages
)

获取生成的新闻稿内容

print(completion.choices[0].message["content"]

部署成功,也能访问,但python调用时候要翻墙

{
"error": {
"message": "Invalid URL (GET /)",
"type": "invalid_request_error",
"param": null,
"code": null
}
}

{
"rewrites": [
{ "source": "/", "destination": "https://api.openai.com" },
{
"source": "/:match*",
"destination": "https://api.openai.com/:match*"
}
],
"github": {
"silent": true
},
"regions": [
"arn1",
"bom1",
"cdg1",
"cle1",
"cpt1",
"dub1",
"fra1",
"gru1",
// "hkg1", ip blocked by openai
"hnd1",
"iad1",
"icn1",
"kix1",
"lhr1",
"pdx1",
"sfo1",
"sin1",
"syd1"
]
}

在 vercel 部署的服务是否支持流式，我看vercel 上好像本身有限制

放在.env中可以么

比如
原始调用方法是

import openai
openai.api_key = api
chat = openai.ChatCompletion.create(
            model="gpt-3.5-turbo",
            messages=messages,
            max_tokens=400,
        )

这段代码是不是修改成

import openai
openai.api_key = api
host = 'YOUR DOMAIN'
url=`https://${host}/v1/chat/completions`
openai.api_base= url
chat = openai.ChatCompletion.create(
            model="gpt-3.5-turbo",
            messages=messages,
            max_tokens=400,
        )

我看到请求都是对话转发，支持embedding 吗？

Reliability: vercel-proxy-openai
·
Failed after 6s

访问也是404

按大神您 的教程，部署了项目，并把自己的一个域名绑上去了，但当我访问openai的特定接口路径（如/v1/models，在浏览器中通过openai.com的域名是可以正常访问的），提示页面没找到。看现象，感觉是反向代理没有生效？请问是什么原因。补充：已经把vercel.json文件提交到github里面。

部署成功了，但是很慢，很多都会报错，如何提高速度

纯小白，自己会照葫芦画瓢。

使用方法
访问 openai api 时，将 "api.openai.com" 换成你的自定义域名，例如
const host = 'YOUR DOMAIN'
const url = https://${host}/v1/chat/completions
在其他 chatgpt 项目中，可以将变量 "OPENAI_API_BASE_URL" 设置为你的域名

我在别人开发好的项目里填入域名都无法成功，和我没操作第一条有关系吗？谢谢

使用python接口访问也是404