Edge Developer Platform
  • AI Gateway
    • Product Introduction
    • Quick Start
    • Development Guide
      • V1
      • V2
        • Overview
        • Introduction to Common Request Headers
        • Tencent-HunYuan
        • Baidu-QianFan
        • Byte-Doubao
        • Ali-TongYi
        • Moonshot-KIMI
        • Minimax
        • Open AI
        • Gemini
    • FAQs
    • Release Notes
Docs Overview/
Edge Developer Platform/
AI Gateway/
Development Guide/
V1/

V1

First, please refer to the Quick Start to create a new AI Gateway and configure the related parameters.
Next, let's introduce the AI Gateway V1 version integration with OPEN AI using a CURL request example.

Request URL

https://ai-gateway-intl.eo-edgefunctions7.com/v1/chat/completions

Request method

POST

Request headers

Parameter name
Required
Type
Description
OE-Key
Yes
string
Gateway API key
OE-Gateway-Name
Yes
string
Gateway name
OE-AI-Provider
Yes
string
AI service provider, fixed value: openai
Authorization
Yes
string
Authentication Key of LLM Service Provider
Content-Type
Yes
string
Fixed value: application/json

Request body

Parameter name
Required
Type
Description
model
Yes
string
Model Name
messages
Yes
array
Message list, each message includes role and content
stream
Yes
boolean
Enable stream processing

Request Body Example
{
  "model": "gpt-3.5-turbo",
  "messages": [
    {
      "role": "user",
      "content": "1+1=?"
    }
  ]
}

Sample response

The response format is JSON, and the specific content depends on the actual API return result.

Sample code for curl request

curl --location --request POST 'https://ai-gateway-intl.eo-edgefunctions7.com/v1/chat/completions' \
--header 'OE-Key: xxxx' \
--header 'OE-Gateway-Name: xxxx' \
--header 'OE-AI-Provider: openai' \
--header 'Authorization: Bearer xxxx' \
--header 'Content-Type: application/json' \
--data-raw '{
    "model":"gpt-3.5-turbo",
    "messages": [
        {
            "role": "user",
            "content": "1+1=?"
        }
    ]
}'

Notes

Note:
1. Please ensure all required parameters are correctly filled out.
2. Adjust the Authorization header based on actual conditions.
3. The messages array in the request body can contain multiple Message Objects, depending on business needs.