We are excited to announce the release of the Griptape Nodes: Google AI Library, a powerful addition to the Griptape Nodes ecosystem that enables easy integration with Google AI services for video and image generation directly within your Griptape workflows. This library unlocks the ability to generate high-quality videos and images using Google’s state-of-the-art AI models, all accessible through intuitive Griptape nodes.
What’s Included?
The Google AI Library for Griptape Nodes currently includes three nodes
- Veo Video Generator: Harness Google’s cutting-edge Veo model to generate videos from text prompts. This node connects to Google’s Vertex AI API to produce high-quality video content based on your creative inputs.
- Video Display: A utility node that dynamically displays video players for the videos generated by the Veo Video Generator. It provides an interactive interface within the Griptape Nodes UI, allowing you to preview and save generated videos seamlessly.
- Imagen Image Generator: Generate stunning images from text prompts using Google’s Imagen text-to-image models. This node also uses Google’s Vertex AI to deliver fast and high-quality image generation and includes support for the latest Imagen 4 models.
Easy Authentication and Setup
The library supports two authentication methods to connect with Google Cloud:
- Service Account File (Recommended for Servers): Create a Google Cloud service account with the Vertex AI User role, download the JSON service accountfile, and provide its path to the nodes. This method is ideal for production environments.
- Application Default Credentials (Recommended for Local Development): Use your local `gcloud` CLI credentials by logging in and setting your project. This method is convenient for development and testing without needing to download and manage service account files.
As usual, the README in the GitHub repo provides detailed step-by-step instructions to help you get started, including on how to create service accounts, enabling the Vertex AI API, and configuring your environment to get started quickly.
Example Workflows
Video Generation
- Add the Veo Video Generator node to your workflow.
- Authenticate using either the service account file or your project ID with default credentials.
- Enter a creative text prompt and specify the number of videos to generate.
- Add the **Video Display** node and connect it to the video generator’s output.
- Run the workflow to generate and instantly preview videos within the Griptape UI.
Image Generation
- Add the Imagen Image Generator node.
- Authenticate similarly as above.
- Provide a text prompt and optionally select the Imagen model and other settings.
- Run the workflow to generate images that appear directly in the node interface.
For this example, I used the the new imagen-4.0-ultra-generate-preview-06-06 model together with the same prompt that I used for the video example above:
A rusty blue pickup truck crafted entirely from textured ice cream, with melting drips and creamy swirls forming its body. The truck's "rust" appears as patches of chocolate fudge and caramel streaks. It races energetically along a street in a futuristic city surrounded by neon illuminated skyscrapers kicking up sprinkles and crushed waffle cone debris. Emphasize the contrast between the cold, creamy truck and the dynamic motion, capturing playful whimsy and detailed textures of ice cream blending with mechanical elements.
Why Use Griptape Nodes: Google AI Library?
This library simplifies integrating Google’s advanced AI video and image generation capabilities into your workflows. By abstracting the complexities of API calls and authentication, it empowers developers and creators with video & image generation at their fingertips.
Get Started Today
The Griptape Nodes: Google AI Library is open source and available now on GitHub. Check out the README for full setup instructions, examples, and detailed documentation.
Unlock new creative possibilities with Google AI and Griptape — start building your AI-powered workflows today!