Generate “High-Quality” 1080p Resolution Video with Google’s AI Model Veo

First teased back in May, Google’s latest generative AI video model officially launched in a private preview. Veo is noteworthy as it was announced around the same time as OpenAI’s Sora (which has perhaps turned heads more than any other AI video model to date), yet has now beat Sora to market.

And with OpenAI still not announcing any release date (or updated plans for that matter) on when Sora might come out, those interested in exploring AI for video are left to utilize what they currently have in front of them.

And with Google Veo, we appear to have a quite sophisticated AI video model that is capable of generating what the company dubs as “high-quality” videos at up to 1080p resolution across a range of different visual and cinematic styles. Let’s take a look at this new text or image prompt based video model and its generative AI possibilities.

Google Veo AI Video Model

Developed by Google’s DeepMind initiative, Veo is here to generate high-quality, high-definition videos based on text or image prompts. The AI model is capable of producing a range of cinematic and visual styles at pretty exceptional speeds.

Veo features an advanced understanding of natural language and visual semantics which should set it apart from some of its main AI video competition and has really been tailored to closely align with the nature of its prompts.

Google’s Veo on Vertex AI further creates footage that’s consistent and coherent so creators can expect people, animals, and objects to move realistically through their generated shots.

Examples of Google Veo AI Video

The real test for any AI is to try it out yourself. However, the second best test is to examine some examples to see if they look like anything you’d like to try to create yourself. Here are some examples shared by Google:

These examples use both prompt types, which include:

Image-to-video: Veo generates videos from existing or AI-generated images. Below are examples of how Veo uses images generated using Imagen 3 (top two images) and real-world images (bottom two images) to create short video clips.

Text-to-video: Below are examples of how Veo uses text to create short video clips.

Pricing and Availability

As mentioned above, Veo has only been released as a private preview and for select business accounts so far. To potentially get started with Veo on Vertex AI you can reach out to your Google Cloud account representative.