Google I/O 2024: Imagen 3 text-to-image AI model revealed, getting enhanced image generation capabilities

Google made several new announcements at its annual developer-focused Google I/O 2024 event. Among the many artificial intelligence (AI)-focused announcements made during the keynote session, one was particularly surprising. The tech giant has unveiled the next generation of its text-to-image AI model, Imagen 3. The new AI model was introduced just months after the release of its predecessor, Imagen 2, which came out in December 2023 and was later upgraded last month. The company said the new model can generate detailed photorealistic images by strictly following the prompt.

Imagen 3 was introduced by Douglas Eck, Senior Research Director at Google DeepMind. Revealing it, he said: “Today I’m so excited to introduce Imagen 3. It’s our most capable imaging model yet. It understands prompts written the way people write. The more creative and detailed you are, the better. Plus, it’s our best model yet for rendering text, which is a challenge for image generation models.”

The AI ​​model’s ability to understand prompts is said to be greatly improved, now allowing it to closely follow the prompt to pick up small details and generate a true image. This also seems to be the general direction for most of the AI-related announcements during the event, as most AI models can now better understand the prompts. Google added that Imagen 3 will be available in multiple versions, where each model is optimized for a specific type of task, which can range from generating quick sketches to creating high-resolution images.

To enable Imagen 3 to pick up small details and specific instructions like camera angles or compositions in long, complex prompts, Google trained the AI ​​model with images that contain detailed descriptions in their captions, allowing it to pick up even smaller shades. It can also generate different textures and render text-based images.

Focusing on safety, every image generated by Imagen 3 will contain the SynthID watermark tag. It embeds a digital watermark directly into the pixels of the image, making it impossible to remove by cropping, sharing or making changes to the image. The AI ​​model is expected to arrive in public review in the coming months. Google is currently working on adding drawing and painting editing options. Imagen 3 is currently available in private preview at ImageFX for select artists. It will soon be available to the tech giant’s enterprise customers.