In August, Google announced that its AI model for generating images would be available to all users of the Gemini service. Now, as promised, the updated Imagen 3 model is accessible to both free and paid users. The model, initially introduced at the Google I/O 2024 event in May, is notable for several key features.
Key Features of Imagen 3
According to the developer, Imagen 3 stands out for three primary reasons:
- Stunning Image Quality: Clear details, vivid colors, and fewer imperfections in generated images.
- Realistic Visual Effects: The AI produces photorealistic visuals with vivid, true-to-life details.
- Diverse Artistic Styles: Users can choose from a wide range of artistic styles, from traditional oil paintings to modern digital art.
How to Use Imagen 3 on Gemini
To create images using Gemini’s AI, users should include keywords like “draw” or “create” in their requests. It’s also possible to specify the style, such as “photorealistic watercolor” or “cartoon illustration,” for more tailored results. The more detailed the request, the more accurate the generated image. After the request is processed, the chatbot will inform the user that the picture was created with the Imagen 3 model. Additionally, any downloaded images will be marked with a SynthID tag for tracking purposes.
Limitations for Free Users
While all users, including those with free accounts, now have access to Imagen 3, there are some limitations, notes NIXsolutions. Free accounts currently cannot generate images that include people, as the system will reject such requests. English-speaking users with Gemini Advanced, Business, and Enterprise subscriptions were the first to gain access to the full capabilities of Imagen 3. For now, these paid subscribers enjoy more comprehensive functionality, but we’ll keep you updated as the service evolves and restrictions may change in the future.