安装¶
输入 [ ]
已复制！
!pip install llama-index-multi-modal-llms-openai
!pip install llama-index-vector-stores-qdrant
!pip install llama_index ftfy regex tqdm
!pip install llama-index-embeddings-clip
!pip install git+https://github.com/openai/CLIP.git
!pip install matplotlib scikit-image
!pip install llama-index-multi-modal-llms-openai !pip install llama-index-vector-stores-qdrant !pip install llama_index ftfy regex tqdm !pip install llama-index-embeddings-clip !pip install git+https://github.com/openai/CLIP.git !pip install matplotlib scikit-image
设置 API 密钥¶
输入 [ ]
已复制！
import os
os.environ["OPENAI_API_KEY"] = "sk-..."
import os os.environ["OPENAI_API_KEY"] = "sk-..."
使用 urls 加载图像¶
输入 [ ]
已复制！
from llama_index.multi_modal_llms.openai import OpenAIMultiModal
from llama_index.core.multi_modal_llms.generic_utils import load_image_urls
image_urls = [
"https://res.cloudinary.com/hello-tickets/image/upload/c_limit,f_auto,q_auto,w_1920/v1640835927/o3pfl41q7m5bj8jardk0.jpg",
]
image_documents = load_image_urls(image_urls)
from llama_index.multi_modal_llms.openai import OpenAIMultiModal from llama_index.core.multi_modal_llms.generic_utils import load_image_urls image_urls = [ "https://res.cloudinary.com/hello-tickets/image/upload/c_limit,f_auto,q_auto,w_1920/v1640835927/o3pfl41q7m5bj8jardk0.jpg", ] image_documents = load_image_urls(image_urls)
使用 GPT-4V 视觉 API 进行查询。¶
输入 [ ]
已复制！
openai_mm_llm = OpenAIMultiModal(
model="gpt-4-vision-preview", max_new_tokens=300
)
openai_mm_llm = OpenAIMultiModal( model="gpt-4-vision-preview", max_new_tokens=300 )
输入 [ ]
已复制！
response = openai_mm_llm.complete(
prompt="Describe the images as an alternative text",
image_documents=image_documents,
)
print(response)
response = openai_mm_llm.complete( prompt="将图像描述为替代文本", image_documents=image_documents, ) print(response)
The image shows the Colosseum in Rome, Italy, illuminated at night with the colors of the Italian flag: green, white, and red. The ancient amphitheater's iconic arches are vividly lit, and the structure stands out against the dark blue evening sky. Some construction or excavation work appears to be taking place in the foreground, and a few people can be seen walking near the site.
从目录加载图像¶
输入 [ ]
已复制！
from pathlib import Path
input_image_path = Path("input_images")
if not input_image_path.exists():
Path.mkdir(input_image_path)
from pathlib import Path input_image_path = Path("input_images") if not input_image_path.exists(): Path.mkdir(input_image_path)
输入 [ ]
已复制！
!wget "https://docs.google.com/uc?export=download&id=1nUhsBRiSWxcVQv8t8Cvvro8HJZ88LCzj" -O ./input_images/long_range_spec.png
!wget "https://docs.google.com/uc?export=download&id=19pLwx0nVqsop7lo0ubUSYTzQfMtKJJtJ" -O ./input_images/model_y.png
!wget "https://docs.google.com/uc?export=download&id=1utu3iD9XEgR5Sb7PrbtMf1qw8T1WdNmF" -O ./input_images/performance_spec.png
!wget "https://docs.google.com/uc?export=download&id=1dpUakWMqaXR4Jjn1kHuZfB0pAXvjn2-i" -O ./input_images/price.png
!wget "https://docs.google.com/uc?export=download&id=1qNeT201QAesnAP5va1ty0Ky5Q_jKkguV" -O ./input_images/real_wheel_spec.png
!wget "https://docs.google.com/uc?export=download&id=1nUhsBRiSWxcVQv8t8Cvvro8HJZ88LCzj" -O ./input_images/long_range_spec.png !wget "https://docs.google.com/uc?export=download&id=19pLwx0nVqsop7lo0ubUSYTzQfMtKJJtJ" -O ./input_images/model_y.png !wget "https://docs.google.com/uc?export=download&id=1utu3iD9XEgR5Sb7PrbtMf1qw8T1WdNmF" -O ./input_images/performance_spec.png !wget "https://docs.google.com/uc?export=download&id=1dpUakWMqaXR4Jjn1kHuZfB0pAXvjn2-i" -O ./input_images/price.png !wget "https://docs.google.com/uc?export=download&id=1qNeT201QAesnAP5va1ty0Ky5Q_jKkguV" -O ./input_images/real_wheel_spec.png
--2024-04-27 20:51:52-- https://docs.google.com/uc?export=download&id=1nUhsBRiSWxcVQv8t8Cvvro8HJZ88LCzj Resolving docs.google.com (docs.google.com)... 172.253.63.113, 172.253.63.139, 172.253.63.100, ... Connecting to docs.google.com (docs.google.com)|172.253.63.113|:443... connected. HTTP request sent, awaiting response... 303 See Other Location: https://drive.usercontent.google.com/download?id=1nUhsBRiSWxcVQv8t8Cvvro8HJZ88LCzj&export=download [following] --2024-04-27 20:51:52-- https://drive.usercontent.google.com/download?id=1nUhsBRiSWxcVQv8t8Cvvro8HJZ88LCzj&export=download Resolving drive.usercontent.google.com (drive.usercontent.google.com)... 172.253.122.132, 2607:f8b0:4004:c09::84 Connecting to drive.usercontent.google.com (drive.usercontent.google.com)|172.253.122.132|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 160684 (157K) [image/png] Saving to: ‘./input_images/long_range_spec.png’ ./input_images/long 100%[===================>] 156.92K --.-KB/s in 0.01s 2024-04-27 20:51:53 (12.3 MB/s) - ‘./input_images/long_range_spec.png’ saved [160684/160684] --2024-04-27 20:51:53-- https://docs.google.com/uc?export=download&id=19pLwx0nVqsop7lo0ubUSYTzQfMtKJJtJ Resolving docs.google.com (docs.google.com)... 172.253.63.113, 172.253.63.139, 172.253.63.100, ... Connecting to docs.google.com (docs.google.com)|172.253.63.113|:443... connected. HTTP request sent, awaiting response... 303 See Other Location: https://drive.usercontent.google.com/download?id=19pLwx0nVqsop7lo0ubUSYTzQfMtKJJtJ&export=download [following] --2024-04-27 20:51:53-- https://drive.usercontent.google.com/download?id=19pLwx0nVqsop7lo0ubUSYTzQfMtKJJtJ&export=download Resolving drive.usercontent.google.com (drive.usercontent.google.com)... 172.253.122.132, 2607:f8b0:4004:c09::84 Connecting to drive.usercontent.google.com (drive.usercontent.google.com)|172.253.122.132|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 1308468 (1.2M) [image/png] Saving to: ‘./input_images/model_y.png’ ./input_images/mode 100%[===================>] 1.25M --.-KB/s in 0.03s 2024-04-27 20:51:55 (38.8 MB/s) - ‘./input_images/model_y.png’ saved [1308468/1308468] --2024-04-27 20:51:55-- https://docs.google.com/uc?export=download&id=1utu3iD9XEgR5Sb7PrbtMf1qw8T1WdNmF Resolving docs.google.com (docs.google.com)... 172.253.63.113, 172.253.63.139, 172.253.63.100, ... Connecting to docs.google.com (docs.google.com)|172.253.63.113|:443... connected. HTTP request sent, awaiting response... 303 See Other Location: https://drive.usercontent.google.com/download?id=1utu3iD9XEgR5Sb7PrbtMf1qw8T1WdNmF&export=download [following] --2024-04-27 20:51:55-- https://drive.usercontent.google.com/download?id=1utu3iD9XEgR5Sb7PrbtMf1qw8T1WdNmF&export=download Resolving drive.usercontent.google.com (drive.usercontent.google.com)... 172.253.122.132, 2607:f8b0:4004:c09::84 Connecting to drive.usercontent.google.com (drive.usercontent.google.com)|172.253.122.132|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 62818 (61K) [image/png] Saving to: ‘./input_images/performance_spec.png’ ./input_images/perf 100%[===================>] 61.35K --.-KB/s in 0.007s 2024-04-27 20:51:55 (8.12 MB/s) - ‘./input_images/performance_spec.png’ saved [62818/62818] --2024-04-27 20:51:55-- https://docs.google.com/uc?export=download&id=1dpUakWMqaXR4Jjn1kHuZfB0pAXvjn2-i Resolving docs.google.com (docs.google.com)... 172.253.63.113, 172.253.63.139, 172.253.63.100, ... Connecting to docs.google.com (docs.google.com)|172.253.63.113|:443... connected. HTTP request sent, awaiting response... 303 See Other Location: https://drive.usercontent.google.com/download?id=1dpUakWMqaXR4Jjn1kHuZfB0pAXvjn2-i&export=download [following] --2024-04-27 20:51:55-- https://drive.usercontent.google.com/download?id=1dpUakWMqaXR4Jjn1kHuZfB0pAXvjn2-i&export=download Resolving drive.usercontent.google.com (drive.usercontent.google.com)... 172.253.122.132, 2607:f8b0:4004:c09::84 Connecting to drive.usercontent.google.com (drive.usercontent.google.com)|172.253.122.132|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 128556 (126K) [image/png] Saving to: ‘./input_images/price.png’ ./input_images/pric 100%[===================>] 125.54K --.-KB/s in 0.01s 2024-04-27 20:51:56 (10.9 MB/s) - ‘./input_images/price.png’ saved [128556/128556] --2024-04-27 20:51:56-- https://docs.google.com/uc?export=download&id=1qNeT201QAesnAP5va1ty0Ky5Q_jKkguV Resolving docs.google.com (docs.google.com)... 172.253.63.113, 172.253.63.139, 172.253.63.100, ... Connecting to docs.google.com (docs.google.com)|172.253.63.113|:443... connected. HTTP request sent, awaiting response... 303 See Other Location: https://drive.usercontent.google.com/download?id=1qNeT201QAesnAP5va1ty0Ky5Q_jKkguV&export=download [following] --2024-04-27 20:51:56-- https://drive.usercontent.google.com/download?id=1qNeT201QAesnAP5va1ty0Ky5Q_jKkguV&export=download Resolving drive.usercontent.google.com (drive.usercontent.google.com)... 172.253.122.132, 2607:f8b0:4004:c09::84 Connecting to drive.usercontent.google.com (drive.usercontent.google.com)|172.253.122.132|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 160998 (157K) [image/png] Saving to: ‘./input_images/real_wheel_spec.png’ ./input_images/real 100%[===================>] 157.22K --.-KB/s in 0.01s 2024-04-27 20:51:57 (12.5 MB/s) - ‘./input_images/real_wheel_spec.png’ saved [160998/160998]
输入 [ ]
已复制！
from PIL import Image
import matplotlib.pyplot as plt
import os
def plot_images(image_paths):
images_shown = 0
plt.figure(figsize=(16, 9))
for img_path in image_paths:
if os.path.isfile(img_path):
image = Image.open(img_path)
plt.subplot(2, 3, images_shown + 1)
plt.imshow(image)
plt.xticks([])
plt.yticks([])
images_shown += 1
if images_shown >= 9:
break
from PIL import Image import matplotlib.pyplot as plt import os def plot_images(image_paths): images_shown = 0 plt.figure(figsize=(16, 9)) for img_path in image_paths: if os.path.isfile(img_path): image = Image.open(img_path) plt.subplot(2, 3, images_shown + 1) plt.imshow(image) plt.xticks([]) plt.yticks([]) images_shown += 1 if images_shown >= 9: break
输入 [ ]
已复制！
image_paths = []
for img_path in os.listdir("./input_images"):
image_paths.append(str(os.path.join("./input_images", img_path)))
plot_images(image_paths)
image_paths = [] for img_path in os.listdir("./input_images"): image_paths.append(str(os.path.join("./input_images", img_path))) plot_images(image_paths)
输入 [ ]
已复制！
from llama_index.multi_modal_llms.openai import OpenAIMultiModal
from llama_index.core import SimpleDirectoryReader
# put your local directore here
image_documents = SimpleDirectoryReader("./input_images").load_data()
from llama_index.multi_modal_llms.openai import OpenAIMultiModal from llama_index.core import SimpleDirectoryReader # put your local directore here image_documents = SimpleDirectoryReader("./input_images").load_data()
输入 [ ]
已复制！
response = openai_mm_llm.complete(
prompt="Describe the images as an alternative text",
image_documents=image_documents,
)
print(response)
response = openai_mm_llm.complete( prompt="将图像描述为替代文本", image_documents=image_documents, ) print(response)
The images provided appear to be screenshots from a website showcasing specifications and details of an electric vehicle, likely from a car manufacturer's website. Here is a description of each image: 1. The first image is a comparison chart of two electric vehicle models, one with Rear-Wheel Drive and the other with Long Range AWD (All-Wheel Drive). It lists various specifications such as battery type, acceleration, range, top speed, cargo volume, seating, wheels, and warranty details. 2. The second image shows a cutaway illustration of an electric vehicle, highlighting its structural components. The image points out the rigid structure and impact protection features of the car's design. 3. The third image is similar to the first one, providing a detailed specification list for the Model Y Performance and Long Range AWD versions of an electric vehicle. It includes information on battery, acceleration, range, top speed, drive type, seating, wheels, and warranty. 4. The fourth image displays pricing information for different variants of the Model Y, including potential savings and federal incentives. It also provides an estimated delivery date and allows the user to enter a postal code for more accurate delivery information. The variants listed are Model Y Rear-Wheel Drive, Model Y Long Range, and Model Y Performance, with their respective prices. 5. The fifth image is another specification chart for an electric vehicle with Rear-Wheel Drive and Long Range AWD options. It lists the same types of specifications as the first image but with different
绘制图像¶
输入 [ ]
已复制！
import requests
def get_wikipedia_images(title):
response = requests.get(
"https://en.wikipedia.org/w/api.php",
params={
"action": "query",
"format": "json",
"titles": title,
"prop": "imageinfo",
"iiprop": "url|dimensions|mime",
"generator": "images",
"gimlimit": "50",
},
).json()
image_urls = []
for page in response["query"]["pages"].values():
if page["imageinfo"][0]["url"].endswith(".jpg") or page["imageinfo"][
0
]["url"].endswith(".png"):
image_urls.append(page["imageinfo"][0]["url"])
return image_urls
import requests def get_wikipedia_images(title): response = requests.get( "https://en.wikipedia.org/w/api.php", params={ "action": "query", "format": "json", "titles": title, "prop": "imageinfo", "iiprop": "url|dimensions|mime", "generator": "images", "gimlimit": "50", }, ).json() image_urls = [] for page in response["query"]["pages"].values(): if page["imageinfo"][0]["url"].endswith(".jpg") or page["imageinfo"][ 0 ]["url"].endswith(".png"): image_urls.append(page["imageinfo"][0]["url"]) return image_urls
从维基百科下载图像和文本¶
输入 [ ]
已复制！
from pathlib import Path
import requests
import urllib.request
import time
image_uuid = 0
# image_metadata_dict stores images metadata including image uuid, filename and path
image_metadata_dict = {}
MAX_IMAGES_PER_WIKI = 15
wiki_titles = {
"Tesla Model Y",
"Tesla Model X",
"Tesla Model 3",
"Tesla Model S",
"Kia EV6",
"BMW i3",
"Audi e-tron",
"Ford Mustang",
"Porsche Taycan",
"Rivian",
"Polestar",
}
data_path = Path("mixed_wiki")
if not data_path.exists():
Path.mkdir(data_path)
for title in wiki_titles:
response = requests.get(
"https://en.wikipedia.org/w/api.php",
params={
"action": "query",
"format": "json",
"titles": title,
"prop": "extracts",
"explaintext": True,
},
).json()
page = next(iter(response["query"]["pages"].values()))
wiki_text = page["extract"]
with open(data_path / f"{title}.txt", "w") as fp:
fp.write(wiki_text)
images_per_wiki = 0
try:
# page_py = wikipedia.page(title)
list_img_urls = get_wikipedia_images(title)
# print(list_img_urls)
for url in list_img_urls:
if (
url.endswith(".jpg")
or url.endswith(".png")
or url.endswith(".svg")
):
image_uuid += 1
# image_file_name = title + "_" + url.split("/")[-1]
urllib.request.urlretrieve(
url, data_path / f"{image_uuid}.jpg"
)
time.sleep(1)
images_per_wiki += 1
# Limit the number of images downloaded per wiki page to 15
if images_per_wiki > MAX_IMAGES_PER_WIKI:
break
except Exception as e:
print(e)
print(
"Number of images found for Wikipedia page: {} are {}".format(
title, images_per_wiki
)
)
continue
from pathlib import Path import requests import urllib.request import time image_uuid = 0 # image_metadata_dict stores images metadata including image uuid, filename and path image_metadata_dict = {} MAX_IMAGES_PER_WIKI = 15 wiki_titles = { "Tesla Model Y", "Tesla Model X", "Tesla Model 3", "Tesla Model S", "Kia EV6", "BMW i3", "Audi e-tron", "Ford Mustang", "Porsche Taycan", "Rivian", "Polestar", } data_path = Path("mixed_wiki") if not data_path.exists(): Path.mkdir(data_path) for title in wiki_titles: response = requests.get( "https://en.wikipedia.org/w/api.php", params={ "action": "query", "format": "json", "titles": title, "prop": "extracts", "explaintext": True, }, ).json() page = next(iter(response["query"]["pages"].values())) wiki_text = page["extract"] with open(data_path / f"{title}.txt", "w") as fp: fp.write(wiki_text) images_per_wiki = 0 try: # page_py = wikipedia.page(title) list_img_urls = get_wikipedia_images(title) # print(list_img_urls) for url in list_img_urls: if ( url.endswith(".jpg") or url.endswith(".png") or url.endswith(".svg") ): image_uuid += 1 # image_file_name = title + "_" + url.split("/")[-1] urllib.request.urlretrieve( url, data_path / f"{image_uuid}.jpg" ) time.sleep(1) images_per_wiki += 1 # Limit the number of images downloaded per wiki page to 15 if images_per_wiki > MAX_IMAGES_PER_WIKI: break except Exception as e: print(e) print( "维基百科页面 '{}' 找到的图像数量为 {}".format( title, images_per_wiki ) ) continue
设置 Qdrant 客户端用于索引¶
输入 [ ]
已复制！
from llama_index.core.indices import MultiModalVectorStoreIndex
from llama_index.vector_stores.qdrant import QdrantVectorStore
from llama_index.core import SimpleDirectoryReader, StorageContext
import qdrant_client
# Create a local Qdrant vector store
client = qdrant_client.QdrantClient(path="qdrant_mm_db")
text_store = QdrantVectorStore(
client=client, collection_name="text_collection"
)
image_store = QdrantVectorStore(
client=client, collection_name="image_collection"
)
storage_context = StorageContext.from_defaults(
vector_store=text_store, image_store=image_store
)
from llama_index.core.indices import MultiModalVectorStoreIndex from llama_index.vector_stores.qdrant import QdrantVectorStore from llama_index.core import SimpleDirectoryReader, StorageContext import qdrant_client # 创建本地 Qdrant 向量存储 client = qdrant_client.QdrantClient(path="qdrant_mm_db") text_store = QdrantVectorStore( client=client, collection_name="text_collection" ) image_store = QdrantVectorStore( client=client, collection_name="image_collection" ) storage_context = StorageContext.from_defaults( vector_store=text_store, image_store=image_store )
加载图像和文本文档¶
输入 [ ]
已复制！
# Create the MultiModal index
documents = SimpleDirectoryReader("./mixed_wiki/").load_data()
# 创建多模态索引 documents = SimpleDirectoryReader("./mixed_wiki/").load_data()
/usr/local/lib/python3.10/dist-packages/PIL/Image.py:3157: DecompressionBombWarning: Image size (101972528 pixels) exceeds limit of 89478485 pixels, could be decompression bomb DOS attack. warnings.warn(
输入 [ ]
已复制！
documents[0]
documents[0]
Out[ ]
ImageDocument(id_='354a7edc-fe0b-4706-9c80-371eacbab684', embedding=None, metadata={'file_path': '/content/mixed_wiki/1.jpg', 'file_name': '1.jpg', 'file_type': 'image/jpeg', 'file_size': 5631170, 'creation_date': '2024-04-27', 'last_modified_date': '2024-04-27'}, excluded_embed_metadata_keys=['file_name', 'file_type', 'file_size', 'creation_date', 'last_modified_date', 'last_accessed_date'], excluded_llm_metadata_keys=['file_name', 'file_type', 'file_size', 'creation_date', 'last_modified_date', 'last_accessed_date'], relationships={}, text='', start_char_idx=None, end_char_idx=None, text_template='{metadata_str}\n\n{content}', metadata_template='{key}: {value}', metadata_seperator='\n', image=None, image_path='/content/mixed_wiki/1.jpg', image_url=None, image_mimetype=None, text_embedding=None)
创建多模态索引¶
输入 [ ]
已复制！
index = MultiModalVectorStoreIndex.from_documents(
documents,
storage_context=storage_context,
)
index = MultiModalVectorStoreIndex.from_documents( documents, storage_context=storage_context, )
100%|████████████████████████████████████████| 338M/338M [00:02<00:00, 125MiB/s] /usr/local/lib/python3.10/dist-packages/PIL/Image.py:3157: DecompressionBombWarning: Image size (101972528 pixels) exceeds limit of 89478485 pixels, could be decompression bomb DOS attack. warnings.warn(
创建查询引擎¶
输入 [ ]
已复制！
from llama_index.core import PromptTemplate
from llama_index.core.query_engine import SimpleMultiModalQueryEngine
qa_tmpl_str = (
"Context information is below.\n"
"---------------------\n"
"{context_str}\n"
"---------------------\n"
"Given the context information and not prior knowledge, "
"answer the query.\n"
"Query: {query_str}\n"
"Answer: "
)
qa_tmpl = PromptTemplate(qa_tmpl_str)
query_engine = index.as_query_engine(
llm=openai_mm_llm, text_qa_template=qa_tmpl
)
from llama_index.core import PromptTemplate from llama_index.core.query_engine import SimpleMultiModalQueryEngine qa_tmpl_str = ( "以下是上下文信息。\n" "---------------------\n" "{context_str}\n" "---------------------\n" "根据上下文信息，而非先验知识，回答查询。\n" "查询：{query_str}\n" "回答：" ) qa_tmpl = PromptTemplate(qa_tmpl_str) query_engine = index.as_query_engine( llm=openai_mm_llm, text_qa_template=qa_tmpl )
查询¶
输入 [ ]
已复制！
query_str = "Tell me more about the Tesla Model X"
response = query_engine.query(query_str)
query_str = "告诉我更多关于特斯拉 Model X 的信息" response = query_engine.query(query_str)
输入 [ ]
已复制！
print(response)
print(response)
The Tesla Model X is a mid-size luxury SUV that is fully electric and known for its unique design and advanced technology features. It was first introduced by Tesla, Inc. in 2015 and has since become one of the most recognizable electric vehicles on the market. One of the most distinctive features of the Model X is its falcon wing doors, which open upwards and allow for easier access to the second and third-row seats. These doors are not only stylish but also functional, as they require less space to open than traditional car doors, making it easier to enter and exit the vehicle in tight parking spaces. The Model X is built on the same platform as the Tesla Model S sedan but is about 10% heavier and shares approximately 30% of its parts with the Model S. It offers a spacious interior with room for up to seven adults and their luggage, thanks to three rows of seating and additional storage in both the front and rear trunks. Safety is a key focus for Tesla, and the Model X is equipped with a number of advanced safety features. According to Tesla CEO Elon Musk, the Model X is extremely safe in terms of frontal and side impact crashes and has a very low risk of rollover. It comes standard with Autopilot, Tesla's advanced driver-assistance system, and has an optional full self-driving system. The vehicle also includes a collision avoidance system with autonomous emergency braking and ultrasound detection to steer the car away from potential threats. In terms of performance, the
显示来源¶
输入 [ ]
已复制！
from llama_index.core.response.notebook_utils import display_source_node
for text_node in response.metadata["text_nodes"]:
display_source_node(text_node, source_length=200)
plot_images(
[n.metadata["file_path"] for n in response.metadata["image_nodes"]]
)
from llama_index.core.response.notebook_utils import display_source_node for text_node in response.metadata["text_nodes"]: display_source_node(text_node, source_length=200) plot_images( [n.metadata["file_path"] for n in response.metadata["image_nodes"]] )
节点 ID: 7ba72e19-49ab-4d1e-9163-af051b680935
相似度 0.9057627702994282
文本: 特斯拉 Model X 是一款自 2015 年起由 Tesla, Inc. 制造的电池电动中型豪华跨界 SUV。该车型基于特斯拉 Model S 的全尺寸轿车平台开发，其显著特点是采用了...
节点 ID: 46ebd206-a399-4ccf-ab8c-f3cbfd1cd5ca
相似度 0.8963384516104667
文本: == 设计 ==
量产版车型于 2015 年 9 月 29 日发布。它配备全景挡风玻璃。据特斯拉 CEO 埃隆·马斯克称，就正面和侧面碰撞而言，它是最安全的 SUV...
