API embedding multimodal

import dashscope
import json
import os
from http import HTTPStatus

# Menggabungkan teks, gambar, dan video menjadi satu embedding terpadu.
# Ideal untuk kasus penggunaan seperti pengambilan lintas-modalitas dan pencarian gambar.
text = "This is a test text for generating a multimodal fused embedding."
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 mencakup teks, gambar, dan video. Atur enable_fusion=True untuk menghasilkan embedding terpadu.
input_data = [
    {"text": text},
    {"image": image},
    {"video": video}
]

resp = dashscope.MultiModalEmbedding.call(
    # Jika variabel lingkungan tidak diatur, berikan Kunci API Model Studio Anda, misalnya api_key="sk-xxx".
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    model="qwen3-vl-embedding",
    input=input_data,
    enable_fusion=True,
    # Opsional: Tentukan dimensi embedding. Nilai valid: 2560, 2048, 1536, 1024, 768, 512, dan 256. Default: 2560.
    # parameters={"dimension": 1024}
)

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

import dashscope
import json
import os
from http import HTTPStatus

# Menggabungkan beberapa gambar produk dan deskripsi menjadi satu embedding.
# Ideal untuk pengambilan semantik komprehensif menggunakan gambar produk multi-sudut dan deskripsi teks.
text = "White sports shoes, lightweight and breathable, suitable for running and daily wear."
image1 = "https://dashscope.oss-cn-beijing.aliyuncs.com/images/256_1.png"
image2 = "https://img.alicdn.com/imgextra/i3/O1CN01rdstgY1uiZWt8gqSL_!!6000000006071-0-tps-1970-356.jpg"

# Teruskan beberapa item gambar dan atur enable_fusion=True untuk menggabungkan semua input menjadi satu embedding.
input_data = [
    {"text": text},
    {"image": image1},
    {"image": image2}
]

resp = dashscope.MultiModalEmbedding.call(
    # Jika variabel lingkungan tidak diatur, berikan Kunci API Model Studio Anda, misalnya api_key="sk-xxx".
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    model="qwen3-vl-embedding",
    input=input_data,
    enable_fusion=True
)

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

import dashscope
import json
import os
from http import HTTPStatus

# Menunjukkan penggunaan parameter res_level (resolusi) dan max_video_frames (frame video).
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"
text = "This is a visual multimodal representation model."

input_data = [
    {"text": text},
    {"image": image},
    {"video": video}
]

resp = dashscope.MultiModalEmbedding.call(
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    model="tongyi-embedding-vision-plus-2026-03-06",
    input=input_data,
    dimension=1152,      # Nilai valid: 1152, 1024, 512, 256, 128, 64
    res_level=1,         # Tingkat resolusi: 0, 1, 2, atau 3. Default: 1.
    max_video_frames=64  # Jumlah maksimum frame video yang diambil. Default: 8. Maksimum: 64.
)

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

import dashscope
import json
import os
from http import HTTPStatus

# Untuk membuat embedding terpadu, tempatkan teks dan gambar dalam objek konten yang sama.
# Model menggabungkan semua input menjadi satu embedding bertipe `fused`.
text = "White sports shoes, lightweight and breathable, suitable for running and daily wear."
image = "https://dashscope.oss-cn-beijing.aliyuncs.com/images/256_1.png"

input_data = [
    {"text": text, "image": image}
]

resp = dashscope.MultiModalEmbedding.call(
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    model="tongyi-embedding-vision-plus-2026-03-06",
    input=input_data,
    dimension=1152
)

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