Google answers DALL-E 3's challenges with Imagen 2
It's based on Google DeepMind's latest generative AI research.
1 min. read
Published on
Share this article
Improve this guide
Read our disclosure page to find out how can you help MSPoweruser sustain the editorial team Read more
Key notes
- Google Cloud is today announcing Imagen 2, the latest iteration of its text-to-image technology.
- The model is trained on a massive dataset of text and images.
- It can generate images with a higher resolution, greater detail, and more realistic lighting than the original Imagen.
Google Cloud is today announcing Imagen 2, the latest iteration of its text-to-image technology. And, from the look of this, this could be a possible competitor to OpenAI’s DALL-E 3 that powers Bing Image Creator.
Imagen 2 is based on Google DeepMind’s latest research in generative AI. The model is trained on a massive dataset of text and images, and it uses this data to learn how to translate text descriptions into corresponding images.
“Imagen 2 on Vertex AI allows our customers to customize and deploy Imagen 2 with intuitive tooling, fully-managed infrastructure, and built-in privacy and safety features,” says the company in the official announcement.
In its demonstration, Google says that Imagen 2 can generate images from a variety of prompts, including descriptions of real-world objects, scenes, and concepts.
According to the tech giant, the model is also able to generate images with a higher resolution, greater detail, and more realistic lighting than the original Imagen. It’s even more accurate when it comes to rendering text, logos, and other objects in images.
“Imagen 2 also includes comprehensive safety filters to help prevent generation of potentially harmful content,” the company reassures.
You can take this model out for a spin on Google Cloud’s website.
