Getty Images Upgrades AI Image Generation Model for Faster and More Accurate Results
In a significant upgrade, Getty Images has announced the release of a new artificial intelligence (AI) image generation model that promises faster generation speeds and higher quality images. The updated model is built on Nvidia’s Edify architecture, which allows for better alignment with prompts, improved 4K upscaling, and increased output speed.
According to Getty Images, the new model can generate four images in just six seconds, a significant improvement over the previous speed. The company also claims that the model can support prompts of up to 250 words in length and can process complex requirements with ease.
The new model also comes with advanced image modification capabilities, allowing users to add or change individual elements, extend the canvas, or remove backgrounds easily. This feature is made possible by the integration of a large language model (LLM) that enables users to fine-tune their image generation needs.
Getty Images has emphasized that the training data for the new model consists solely of high-resolution, licensed, or owned images and metadata from its vast creative library. The company has also assured that no data or images were scraped from the internet, generated synthetically, or sourced from other generators.
Moreover, the image generator has been trained to respect intellectual property and artist rights. As a result, prompts requesting images of identifiable people, protected locations, trademarks, or brands will not be fulfilled.
The updated AI model will power Getty Images’ AI tools, including Generative AI by Getty Images and Generative AI by iStock. The company has announced that the new model is now available for testing on its platform, marking a significant milestone in the development of AI-powered image generation technology.
