RzenEmbed API

RzenEmbed 是一个兼容 OpenAI API 格式的多模态 Embedding 服务，支持将文本、图像、视频及图文混排内容转换为统一语义空间中的高维向量表示。

接入说明

• 仅支持 POST /v1/embeddings
• 不支持 /models/interface
• 模型名固定为 RzenEmbed
• 与普通 /v1/embeddings 模型的区别在于：输入内容通过 texts、images、video、instruction、request_id 等字段传入，而不是只依赖简单的 input 字符串
• 如果使用 OpenAI SDK，非标准字段需放在 extra_body 中

请求地址

• Method: POST
• URL: https://api.research.360.cn/v1/embeddings

请求头

Header	必填	说明
Authorization	是	Bearer <token>
accept	是	固定 application/json
Content-Type	是	固定 application/json

请求参数

参数	类型	必填	说明
model	string	是	固定传 RzenEmbed
input	array	否	OpenAI SDK 占位参数，可传 []
dimensions	int	否	默认 2048，仅支持 256 / 512 / 768 / 1024 / 1536 / 2048
request_id	string	否	请求唯一标识，不传则自动生成
instruction	string	否	指令文本，默认 Represent the user's input.
texts	array[string]	否	多模态文本输入，仅允许 1 条
images	array[string]	否	多模态图片输入，仅允许 1 张，支持 URL 或 Base64
video	array[string]	否	视频帧输入，支持 2 到 64 张图片，支持 URL 或 Base64

参数约束

• texts、images、video 至少提供一种
• texts 最多 1 条
• images 最多 1 张
• video 需要 2 到 64 张图片帧
• OpenAI SDK 调用时，request_id、instruction、texts、images、video 需通过 extra_body 传入

响应格式

响应遵循 OpenAI Embedding 对象结构，并额外返回 message、response_status、request_id 字段。

• message = success 表示请求正常
• response_status = 0 表示服务成功
• request_id 为本次请求唯一标识

{
  "data": [
    {
      "embedding": [
        -0.07196125388145447,
        -0.0329570434987545,
        0.0068786488845944405
      ],
      "index": 0,
      "object": "embedding"
    }
  ],
  "model": "RzenEmbed",
  "object": "list",
  "usage": {
    "prompt_tokens": 1761,
    "total_tokens": 1761
  },
  "message": "success",
  "response_status": 0,
  "request_id": "b978be6f-151d-4758-9fdc-b25e19019801"
}

调用示例

1. 单条文本 Embedding

curl --location 'https://api.research.360.cn/v1/embeddings'   --header 'accept: application/json'   --header 'Content-Type: application/json'   --header 'Authorization: Bearer your_key'   --data '{
    "model": "RzenEmbed",
    "request_id": "5fe3f69f-e4df-47c9-8ceb-c5c3cbe65daf",
    "dimensions": 2048,
    "instruction": "Represent the user''s input.",
    "texts": ["two apples"]
  }'

import uuid
from openai import OpenAI

client = OpenAI(base_url="https://api.research.360.cn/v1", api_key="your_key")

response = client.embeddings.create(
    model="RzenEmbed",
    input=[],
    dimensions=2048,
    extra_body={
        "request_id": str(uuid.uuid4()),
        "texts": ["two apples"],
        "instruction": "Represent the user's input."
    }
)
print(response)

2. 单张图像 Embedding

curl --location 'https://api.research.360.cn/v1/embeddings'   --header 'accept: application/json'   --header 'Content-Type: application/json'   --header 'Authorization: Bearer your_key'   --data '{
    "model": "RzenEmbed",
    "request_id": "5fe3f69f-e4df-47c9-8ceb-c5c3cbe65daf",
    "dimensions": 2048,
    "instruction": "Find me an everyday image that matches the given caption:",
    "images": ["https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg"]
  }'

import uuid
from openai import OpenAI

client = OpenAI(base_url="https://api.research.360.cn/v1", api_key="your_key")

response = client.embeddings.create(
    model="RzenEmbed",
    input=[],
    dimensions=2048,
    extra_body={
        "request_id": str(uuid.uuid4()),
        "instruction": "Find me an everyday image that matches the given caption:",
        "images": ["https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg"]
    }
)
print(response)

3. 单个视频 Embedding

curl --location 'https://api.research.360.cn/v1/embeddings'   --header 'accept: application/json'   --header 'Content-Type: application/json'   --header 'Authorization: Bearer your_key'   --data '{
    "model": "RzenEmbed",
    "request_id": "5fe3f69f-e4df-47c9-8ceb-c5c3cbe65daf",
    "dimensions": 2048,
    "instruction": "Find me an everyday video that matches the given caption:",
    "video": [
      "https://p0.ssl.qhimg.com/d/inn/20b6511288e3/Catch_med_5/10000.jpg",
      "https://p0.ssl.qhimg.com/d/inn/20b6511288e3/Catch_med_5/10001.jpg"
    ]
  }'

import uuid
from openai import OpenAI

client = OpenAI(base_url="https://api.research.360.cn/v1", api_key="your_key")

response = client.embeddings.create(
    model="RzenEmbed",
    input=[],
    dimensions=2048,
    extra_body={
        "request_id": str(uuid.uuid4()),
        "instruction": "Find me an everyday video that matches the given caption:",
        "video": [
            "https://p0.ssl.qhimg.com/d/inn/20b6511288e3/Catch_med_5/10000.jpg",
            "https://p0.ssl.qhimg.com/d/inn/20b6511288e3/Catch_med_5/10001.jpg"
        ]
    }
)
print(response)

4. 图文混合 Embedding

curl --location 'https://api.research.360.cn/v1/embeddings'   --header 'accept: application/json'   --header 'Content-Type: application/json'   --header 'Authorization: Bearer your_key'   --data '{
    "model": "RzenEmbed",
    "request_id": "5fe3f69f-e4df-47c9-8ceb-c5c3cbe65daf",
    "dimensions": 2048,
    "instruction": "Find me an everyday image that matches the given caption:",
    "texts": ["two apples"],
    "images": ["https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg"]
  }'

import uuid
from openai import OpenAI

client = OpenAI(base_url="https://api.research.360.cn/v1", api_key="your_key")

response = client.embeddings.create(
    model="RzenEmbed",
    input=[],
    dimensions=2048,
    extra_body={
        "request_id": str(uuid.uuid4()),
        "instruction": "Find me an everyday image that matches the given caption:",
        "texts": ["two apples"],
        "images": ["https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg"]
    }
)
print(response)