Imagen 3 Arrives In Google Docs
Users can now generate AI images in Google Docs. The image generator will be powered by Imagen 3.
In an exciting move to keep up with Microsoft’s integration of AI image generation into its Office apps, Google has introduced its own image-generation feature in Google Docs. This new tool is powered by Gemini’s Imagen 3.
The feature is currently available for paid workspace accounts, including those with the Gemini Business, Enterprise, Education, Education Premium, or Google One AI Premium add-ons.
To get started, open a blank Doc file, and under the Insert toolbar, select Image, and in the sub-menu you’ll notice a new “Help me create an image” option.
A panel appears on the right side of the screen. This is where you type in your image prompt and adjust key settings such as aspect ratio and style.
For aspect ratios, you can choose square, horizontal, or vertical to suit the layout of your flier, brochure, article cover, or whatever project you’re working on.
Let’s try out the example prompt below:
Prompt: A cute bear made out of cotton candy
Before generating, ensure your settings are optimized. For this example, I set the aspect ratio to wide and the style to photography.
After entering the prompt, the tool takes around 20 seconds to process, producing four variations.
From here, you can select the version you like best, and it will be instantly inserted into your document.
This is great if you usually work with documents with custom images.
Previously, sourcing images required toggling between tabs or using third-party tools. Now, everything you need is built directly into Google Docs, streamlining your workflow.
To maximize the quality of your AI-generated images, Google offers the following tips:
Be specific in your prompts. Consider the subject, setting, distance, materials, and background. Example: A close-up of a boat made out of wood on a lake surrounded by trees at sunrise.
Avoid figurative language or overly abstract concepts. AI struggles with metaphors or vague descriptions.
Image Styles
Aside from the default “no style,” there are currently five image styles to choose from.
Photography
Vector Art
Sketch
Watercolor
Cyberpunk
Additionally, there’s an “I’m feeling lucky” option that randomly selects one of these styles for you.
Here’s a closer look at some styles with sample prompts:
Image style: Vector art
Prompt: A teddy bear dressed up as a warrior on top of mountain, cinematic photography sunset lighting
Image style: Sketch
Prompt: Dreamy photograph of a sunlit tropical rainforest with colorful flowers, a waterfall cascading down a cliff in the background
Style: Cyberpunk
Prompt: 3D render surrealist Roman architecture with geometric pink arches
Each style gives a unique vibe, so you can pick what works best for your project. One thing I would’ve loved is the ability to upload your own style, and the AI would copy that style to the generated image.
Creating A Cover Image
If you want a professional-looking cover image, the new image generator tool also supports this. Cover images are only available in pageless mode. When you change your document to Pages mode, the cover image is hidden.
To set the document to pageless mode, go to File > Page Setup and then select the Pageless tab and click OK.
Once you’ve switched modes, a new Cover image option appears under the Insert menu. From here, you can generate your cover image.
Alternatively, you can bring up the image generator tool by typing @cover image
on the document.
Now you’re ready to generate images. Here’s an example:
Prompt: A majestic snow-capped mountain range, dreamlike
Cover images automatically span the full width of the document, so there’s no need to adjust the aspect ratio.
In-app AI Text Generator
Aside from the new image generator, Google also added a new AI-powered text generator called “Help me write.”
For instance, if you wanted to write a children’s story, you could now create it directly in Google Docs instead of relying on a chatbot like Gemini.
Prompt: Help me write a short story about a boy and a dog who got lost in the woods
And just like that, you have a story. Now, to give it a cover image, simply create it using the new image generator.
Prompt: A boy and a dog playing in the jungle
From my experience, the feature seems to be a bit unstable because I keep getting this error: “We’re still learning and can’t help with that. Try something else.”
In order to avoid such errors, try modifying the prompt or refreshing the page and trying again.
Final Thoughts
I’m really excited to see this feature in action, especially with how much better Imagen 3 seems compared to the older versions.
As someone who spends a lot of time writing drafts in Google Docs, I often need a quick cover image to make the document feel more complete. Before this, I’d grab something from Unsplash or even Google Images just as a placeholder. But now, being able to generate an AI image right inside Docs makes things so much easier.
The images themselves are decent—good enough for drafts or quick projects. Of course, they’re not on the same level as MidJourney or Flux Ultra, but for something built right into Google Docs, it’s a good start.
I’m curious to hear what you think. What kind of projects or workflows do you see this being useful for?
Hi there, thanks for making it to the end of this post! My name is Jim, and I’m an AI enthusiast passionate about exploring the latest news, guides, and insights in the world of generative AI. If you’ve enjoyed this content and would like to support my work, consider becoming a paid subscriber to my publication. Your support means a lot!