探索 Gemini API 的视觉功能

from google import genai
from google.genai import types

import PIL.Image

image = PIL.Image.open('/path/to/image.png')

client = genai.Client(api_key="GEMINI_API_KEY")
response = client.models.generate_content(
    model="gemini-2.0-flash",
    contents=["What is this image?", image])

print(response.text)

from google import genai
from google.genai import types

import requests

image_path = "https://goo.gle/instrument-img"
image = requests.get(image_path)

client = genai.Client(api_key="GEMINI_API_KEY")
response = client.models.generate_content(
    model="gemini-2.0-flash-exp",
    contents=["What is this image?",
              types.Part.from_bytes(image.content, "image/jpeg")])

print(response.text)

from google import genai
from google.genai import types

import pathlib
import PIL.Image

image_path_1 = "path/to/your/image1.jpeg"  # Replace with the actual path to your first image
image_path_2 = "path/to/your/image2.jpeg" # Replace with the actual path to your second image

image_url_1 = "https://goo.gle/instrument-img" # Replace with the actual URL to your third image

pil_image = PIL.Image.open(image_path_1)

b64_image = types.Part.from_bytes(
    pathlib.Path(image_path_2).read_bytes(), "image/jpeg")

downloaded_image = requests.get(image_url_1)

client = genai.Client(api_key="GEMINI_API_KEY")
response = client.models.generate_content(
    model="gemini-2.0-flash-exp",
    contents=["What do these images have in common?",
              pil_image, b64_image, downloaded_image])

print(response.text)

from google import genai

client = genai.Client(api_key="GEMINI_API_KEY")

img_path = "/path/to/Cajun_instruments.jpg"
file_ref = client.files.upload(path=img_path)
print(f'{file_ref=}')

client = genai.Client(api_key="GEMINI_API_KEY")
response = client.models.generate_content(
    model="gemini-2.0-flash-exp",
    contents=["What can you tell me about these instruments?",
              file_ref])

print(response.text)

from google import genai

client = genai.Client(api_key="GEMINI_API_KEY")

prompt = (
  "Return a bounding box for each of the objects in this image "
  "in [ymin, xmin, ymax, xmax] format.")

response = client.models.generate_content(
  model="gemini-1.5-pro",
  contents=[sample_file_1, prompt])

print(response.text)

wget https://storage.googleapis.com/generativeai-downloads/images/GreatRedSpot.mp4

from google import genai

client = genai.Client(api_key="GEMINI_API_KEY")

print("Uploading file...")
video_file = client.files.upload(path="GreatRedSpot.mp4")
print(f"Completed upload: {video_file.uri}")

import time

# Check whether the file is ready to be used.
while video_file.state.name == "PROCESSING":
    print('.', end='')
    time.sleep(1)
    video_file = client.files.get(name=video_file.name)

if video_file.state.name == "FAILED":
  raise ValueError(video_file.state.name)

print('Done')

from IPython.display import Markdown

# Pass the video file reference like any other media part.
response = client.models.generate_content(
    model="gemini-1.5-pro",
    contents=[
        video_file,
        "Summarize this video. Then create a quiz with answer key "
        "based on the information in the video."])

# Print the response, rendering any Markdown
Markdown(response.text)

prompt = "What are the examples given at 01:05 and 01:19 supposed to show us?"

response = client.models.generate_content(
    model="gemini-1.5-pro",
    contents=[video_file, prompt])

print(response.text)

prompt = (
    "Transcribe the audio from this video, giving timestamps for "
    "salient events in the video. Also provide visual descriptions.")

response = client.models.generate_content(
    model="gemini-1.5-pro",
    contents=[video_file, prompt])

print(response.text)

from google import genai

client = genai.Client(api_key="GEMINI_API_KEY")

print('My files:')
for f in client.files.list():
  print(" ", f'{f.name}: {f.uri}')

from google import genai

client = genai.Client(api_key="GEMINI_API_KEY")

# Upload a file
poem_file = client.files.upload(path="poem.txt")

# Files will auto-delete after a period.
print(poem_file.expiration_time)

# Or they can be deleted explicitly.
dr = client.files.delete(name=poem_file.name)

try:
  client.models.generate_content(
      model="gemini-2.0-flash-exp",
      contents=['Finish this poem:', poem_file])
except genai.errors.ClientError as e:
  print(e.code)  # 403
  print(e.status)  # PERMISSION_DENIED
  print(e.message)  # You do not have permission to access the File .. or it may not exist.