We’re thrilled to announce that Veo 2, our state-of-the-art video generation model, is now available for developers to integrate into their applications. You can experiment with its capabilities in Google AI Studio and then start building on a paid tier in the Gemini API.
Veo 2 represents a new frontier in turning text, images, or both into video. It excels at interpreting both simple and complex instructions from text or image prompts, generating eight second video clips that accurately simulate real-world physics and capture a diverse spectrum of visual and cinematic styles.
Veo 2 empowers developers to generate eight second videos directly within their applications from both text and image prompts:
The easiest way to start experimenting and exploring Veo 2’s capabilities is directly within Google AI Studio. You will be able to test prompts, adjust parameters like aspect ratio and duration, then immediately see the generated video results.
Use Google AI Studio to familiarize yourself with Veo 2’s potential. Once you’re ready to integrate Veo 2’s power directly into your own applications and workflow, you can leverage the Gemini API.
import time
from google import genai
from google.genai import types
client = genai.Client()
operation = client.models.generate_videos(
model="veo-2.0-generate-001",
prompt="Panning wide shot of a calico kitten sleeping in the sunshine",
config=types.GenerateVideosConfig(
person_generation="allow_adult",
aspect_ratio="16:9",
),
)
while not operation.done:
time.sleep(20)
operation = client.operations.get(operation)
for n, generated_video in enumerate(operation.response.generated_videos):
client.files.download(file=generated_video.video)
generated_video.video.save(f"video{n}.mp4") # save the video
Generating stunning videos with Veo 2 hinges on your ability to communicate your vision clearly and effectively. Think of your prompt as a set of instructions – the more detailed and precise you are, the closer the final product will be to what you imagined. The key elements are: clarity, detail, and visual keywords. Let’s break this down with examples.
Clarity: Avoid vague terms and general descriptions.
Detail: The more information you provide, the richer and more nuanced the generated video will be.
Consider elements like:
Let’s review a couple examples.
Let's say you want a video showcasing a new perfume bottle. Here's how you can build up a detailed prompt:
This works by detailing the Subject [faceted crystal bottle, rose gold accents, marble surface], Action [drops rolls down the side], Lighting [soft,diffused light], Camera Angle [close-up shot], and Style [elegant, luxurious]
Example 2: Get more precise by using Image-to-Video capabilities.
Use Image-to-Video capabilities to showcase an existing product following your style and aesthetic. Upload an existing image or create one with Imagen:
Prompt:
Create a luxurious promotional video showcasing a perfume bottle. Begin with a tight close-up dolly left shot, focusing on the faceted cap of a clear glass perfume bottle filled with amber liquid. Water droplets subtly cling to the glass. The bottle rests on a clean, white marble bathroom countertop. Soft, natural light streams in from a window in the background, illuminating the scene. Eucalyptus leaves and natural wood fragrance diffuser sticks are subtly arranged around the bottle. The overall mood is elegant, fresh, and sophisticated.
Video output:
By mastering these principles, you'll be well on your way to crafting prompts that unlock the full potential of Veo 2 and bring your creative visions to life. Remember to iterate and refine your prompts based on the results you get – experimentation is key!
To illustrate the transformative potential of Veo 2, let's look at how developers are already leveraging it to build next-generation creative tools.
AlphaWave helps fashion and retail brands scale their content production using AI. Their core tool, AlphaFrame, automates the creation of high-performing marketing videos, solving the challenge of quickly and cost-effectively producing engaging, conversion-focused content for product drops and promotions.
By integrating Veo 2, AlphaWave can now generate polished, brand-aligned videos in minutes from simple text prompts or existing static assets like product images. This enables their clients to rapidly test ad variations, turn static catalogs into dynamic motion content, and empower brands with limited resources to access quality video production, ultimately making them more agile and competitive.
In the example below, AlphaWave has taken the static Pixel product images and turned it into a dynamic marketing video.
Trakto helps teams scale the creation of high-quality marketing assets with its creative automation platform. To accelerate video production, their AI-powered Trakto Director feature transforms simple prompts into complete, editable commercials.
After Gemini Flash scripts scenes and Imagen creates storyboard visuals, Veo 2 generates the final video. For Trakto, Veo 2 is crucial, delivering the temporal consistency, creative understanding, format flexibility, and polished output needed to rapidly turn ideas into high-quality, adaptable video content, significantly streamlining the path from concept to campaign-ready asset.
Veo 2 is ready to revolutionize how you create and integrate video content. Dive deeper and start building:
We can't wait to see what you create with Veo 2!
Gemini 2.5 Flash and Pro, Live API, and Veo 2 in the Gemini API
Apigee announces general availability of APIM Extension Processor
Agent Development Kit: Making it easy to build multi-agent applications
Unlock your potential: Discover the enhanced Google Developer Program