Veo 3: The Definitive Guide to Google's Newest AI Video Generation Model

from google.generativeai import GenerativeModel
from google.veo import VideoGenerator

# Step 1: Use Gemini 3 to plan the video
gemini = GenerativeModel('gemini-3-pro-preview-11-2025')

planning_prompt = """
I want to create a 90-second product demo video for our new 
electric bike. It should feel modern and energetic. 
Break this into 4-5 shots with detailed descriptions.
"""

storyboard = gemini.generate_content(planning_prompt)
print("Storyboard:", storyboard.text)

# Step 2: Generate each shot with Veo 3
veo = VideoGenerator('veo-3')
shots = []

for shot_description in storyboard.shots:
    video_clip = veo.generate(
        prompt=shot_description,
        duration=15,  # seconds
        style="cinematic",
        resolution="1080p",
        audio=True
    )
    shots.append(video_clip)

# Step 3: Ask Gemini 3 to create transitions
transitions = gemini.generate_content(
    f"Suggest smooth transitions between these shots: {shots}"
)

# Step 4: Combine into final video
final_video = veo.combine_clips(
    shots=shots,
    transitions=transitions.suggestions,
    background_music="upbeat-electronic"
)

final_video.export("electric_bike_demo.mp4")

# Install Google Video SDK

# Generate your first video

from google.veo import VideoGenerator

# Initialize Veo 3
veo = VideoGenerator('veo-3')

# Create a simple video
video = veo.generate(
    prompt="A serene beach at sunset, waves gently lapping the shore, "
           "cinematic wide shot, golden hour lighting",
    duration=15,  # seconds
    resolution="1080p",
    style="cinematic"
)

# Save the result
video.save("beach_sunset.mp4")
print(f"Video generated successfully: {video.url}")