Alibaba Cloud Model Studio：Detail API penyematan multimodal

Hasilkan penyematan gambar

import dashscope
import json
from http import HTTPStatus
# Ganti URL dengan URL gambar Anda.
image = "https://dashscope.oss-cn-beijing.aliyuncs.com/images/256_1.png"
input = [{'image': image}]
# Panggil API model.
resp = dashscope.MultiModalEmbedding.call(
    model="tongyi-embedding-vision-plus",
    input=input
)

if resp.status_code == HTTPStatus.OK:
    result = {
        "status_code": resp.status_code,
        "request_id": getattr(resp, "request_id", ""),
        "code": getattr(resp, "code", ""),
        "message": getattr(resp, "message", ""),
        "output": resp.output,
        "usage": resp.usage
    }
    print(json.dumps(result, ensure_ascii=False, indent=4))

import dashscope
import base64
import json
from http import HTTPStatus
# Baca gambar dan konversi ke Base64. Ganti xxx.png dengan nama file atau path gambar Anda.
image_path = "xxx.png"
with open(image_path, "rb") as image_file:
    # Baca file dan konversi ke Base64.
    base64_image = base64.b64encode(image_file.read()).decode('utf-8')
# Tetapkan format gambar.
image_format = "png"  # Ubah sesuai format aktual, misalnya jpg atau bmp.
image_data = f"data:image/{image_format};base64,{base64_image}"
# Data input.
input = [{'image': image_data}]

# Panggil API model.
resp = dashscope.MultiModalEmbedding.call(
    model="tongyi-embedding-vision-plus",
    input=input
)
if resp.status_code == HTTPStatus.OK:
    result = {
        "status_code": resp.status_code,
        "request_id": getattr(resp, "request_id", ""),
        "code": getattr(resp, "code", ""),
        "message": getattr(resp, "message", ""),
        "output": resp.output,
        "usage": resp.usage
    }
    print(json.dumps(result, ensure_ascii=False, indent=4))

Hasilkan penyematan video

Model penyematan multimodal saat ini hanya mendukung input file video dalam bentuk URL. Pengiriman video lokal secara langsung tidak didukung.

import dashscope
import json
from http import HTTPStatus
# Ganti URL dengan URL video Anda.
video = "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20250107/lbcemt/new+video.mp4"
input = [{'video': video}]
# Panggil API model.
resp = dashscope.MultiModalEmbedding.call(
    model="tongyi-embedding-vision-plus",
    input=input
)

if resp.status_code == HTTPStatus.OK:
    result = {
        "status_code": resp.status_code,
        "request_id": getattr(resp, "request_id", ""),
        "code": getattr(resp, "code", ""),
        "message": getattr(resp, "message", ""),
        "output": resp.output,
        "usage": resp.usage
    }
    print(json.dumps(result, ensure_ascii=False, indent=4))
    

Hasilkan penyematan teks

import dashscope
import json
from http import HTTPStatus

text = "Contoh model representasi multimodal umum"
input = [{'text': text}]
# Panggil API model.
resp = dashscope.MultiModalEmbedding.call(
    model="tongyi-embedding-vision-plus",
    input=input
)

if resp.status_code == HTTPStatus.OK:
    result = {
        "status_code": resp.status_code,
        "request_id": getattr(resp, "request_id", ""),
        "code": getattr(resp, "code", ""),
        "message": getattr(resp, "message", ""),
        "output": resp.output,
        "usage": resp.usage
    }
    print(json.dumps(result, ensure_ascii=False, indent=4))

Hasilkan penyematan terpadu

import dashscope
import json
import os
from http import HTTPStatus

# Penyematan multimodal terpadu: Menggabungkan teks, gambar, dan video menjadi satu penyematan terpadu.
# Cocok untuk skenario seperti pengambilan lintas-modalitas dan pencarian gambar.
text = "Ini adalah teks uji untuk menghasilkan penyematan multimodal terpadu."
image = "https://dashscope.oss-cn-beijing.aliyuncs.com/images/256_1.png"
video = "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20250107/lbcemt/new+video.mp4"

# Input berisi teks, gambar, dan video. Model akan menggabungkannya menjadi satu penyematan terpadu.
input_data = [
    {
        "text": text,
        "image": image,
        "video": video
    }
]

# Gunakan qwen3-vl-embedding untuk menghasilkan penyematan terpadu.
resp = dashscope.MultiModalEmbedding.call(
    # Jika Anda belum mengonfigurasi variabel lingkungan, ganti baris berikut dengan Kunci API Model Studio Anda: api_key="sk-xxx",
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    model="qwen3-vl-embedding",
    input=input_data,
    # Parameter opsional: Tentukan dimensi vektor (nilai yang didukung: 2560, 2048, 1536, 1024, 768, 512, 256; default: 2560)
    # parameters={"dimension": 1024}
)

print(json.dumps(resp, ensure_ascii=False, indent=4))