Google unveils Imagen 3: Its most advanced text-to-image model

Google unveils Imagen 3: Its most advanced text-to-image model is profiled by BTW Media because published evidence links it to internet infrastructure, governance, operational dependencies, or market visibility.

• Imagen 3 is a photorealistic image generator that excels with detailed, creative prompts.
• Google unveils Imagen 3 as part of Google’s strategy to stay ahead in the competitive AI image generation market.
• Imagen 3 integration with Vertex AI is anticipated to bring advanced image generation capabilities to a wider audience, enhancing various applications across industries.

Google has introduced Imagen 3, its highest-quality text-to-image model to date, available in private preview. This new model promises to deliver unprecedented detail, enhanced natural language understanding, and improved text rendering capabilities.

Enhancements and competitive edges

Douglas Eck, Senior Research Director of Google DeepMind, highlighted the advancements of Imagen 3 during the company’s I/O developer conference. “It’s our most capable image generation model yet,” Eck stated. “Imagen 3 is more photorealistic, with richer details and fewer visual artefacts or distorted images. It understands prompts written the way people write—the more creative and detailed you are, the better. And Imagen 3 remembers to incorporate small details in longer prompts. Plus, this is our best model yet for rendering text, which has been a challenge for image generation models.”

Also read: Google is adding its Gemini Nano AI model to desktop Chrome

The launch of Imagen 3 comes six months after the release of Imagen 2 on Vertex AI. This move is part of Google’s strategy to stay ahead in the competitive AI landscape, where rivals such as OpenAI’s DALL-E, Midjourney, Adobe’s Firefly, Meta’s AI, and Microsoft’s Designer are also innovating in image generation technology. Google’s recent rollout of text-to-live capabilities for its models in April further demonstrates its commitment to leading the AI image generation market.

Imagen 3 is currently available in private preview for select users, with a broader rollout expected as developers can sign up for the waitlist. The integration with Vertex AI is anticipated to bring advanced image generation capabilities to a wider audience, enhancing various applications across industries.

Also read: Google launches Trillium AI chip that’s five times faster

Addressing controversies

Despite its advancements, Google has faced challenges in this space. Earlier this year, the company was criticised for its AI-generated images depicting historical figures and events inaccurately, which led to accusations of the AI being “too woke.” Google apologized for these incidents, with CEO Sundar Pichai reiterating the company’s commitment to improving its models during an interview with Bloomberg’s Emily Chang.