Create Embedding

📘 API Introduction (Embeddings API)

The Embeddings API is used to convert text into high-dimensional vectors (vector representation), which are widely applied in scenarios such as semantic search, similarity calculation, clustering, and recommendation systems.

This endpoint maps text onto a fixed-dimension floating-point array (vector) to serve as a machine-understandable semantic representation.

🎯 Applicable Scenarios
🔎 Semantic Search

Perform similarity matching between user queries and document vectors (e.g., knowledge base retrieval)

🤖 RAG (Retrieval-Augmented Generation)

Combine with vector databases to achieve context-enhanced Q&A

📊 Text Clustering

Automatically classify text based on semantic similarity

🧠 Recommendation Systems

Execute recommendation ranking based on the semantic similarity of content

🔐 Authentication Method

All requests must be authenticated via a Bearer Token.

Header Example
Authorization: Bearer sk-xxxxxxxxxx
Content-Type: application/json

请求示例代码

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

返回响应

🟢200

application/json

Successful Response

Bodyapplication/json

示例

{
    "object": "list",
    "data": [
        {
            "object": "embedding",
            "index": 0,
            "embedding": [
                0.0023063174,
                -0.009361042,
                0.015764702,
                -0.007297516,
                0.004337311,
                -0.00058698654,
                0.0006623268,
                0.02305603,
                -0.005214691,
                -0.01725769,
                -0.005870819,
                0.015945435,
                -0.015335083,
                0.0005760193,
                0.025787354,
                0.014816284,
                -0.023544312,
                -0.015090942,
                0.006893158,
                -0.008895874,
                -0.016586304,
                0.0035057068
            ]
        }
    ],
    "model": "text-embedding-3-large",
    "usage": {
        "prompt_tokens": 16,
        "total_tokens": 16
    }
}

请求参数

请求示例代码

返回响应