Quick Links
If Midjourney, DALL-E, or Stable Diffusion are too advanced or expensive for your needs, you can use Google Bard for AI image generation. Below, we explain how to generate AI images with Google Bard, along with a few tips for getting the best results.
How to Generate AI Images Using Google Bard
Generating AI images withGoogle Bardis quite simple (and completely free). You can ask Bard to generate an image of anything you’d like, but you have to make it clear that you want a generated image. Otherwise, you might end up getting an explanation of your prompt accompanied by an existing image.
Use words like “generate” or “create” at the beginning of your prompt, and Bard will respond with two unique AI-generated images.
If you don’t like the results, clickGenerate morebelow the images, and Bard will respond with two more images. Alternatively, you can enter another prompt, giving Bard additional instructions about the kind of image you want.
Once Bard has generated the perfect image, you may hover your mouse over it and click thedownward arrowicon to download it. You can also download both images together by clicking on theshareicon and selectingDownload all images.
As mentioned inthe Google blog post, Bard uses Google’s AI text-to-image diffusion model called Imagen 2 for image generation. If you’re interested, you canaccess Imagen 2 through a free Google Cloud account.
To helpidentify AI images, Bard adds a SynthID watermark, which is unnoticeable to the human eye. The watermark can be detected viaSynthIDand remains unaffected by compression or filters.
At the time of writing, Bard can only generate images using English prompts and isn’t available in the European Economic Area, the UK, or Switzerland.
Quick Tips for Generating Images With Google Bard
While using Bard is pretty straightforward, getting it to generate good images can be challenging. If you’re struggling with that part, here are a few quick tips:
1. Be Descriptive and Clear
The more descriptive your prompts are, the more likely you are to get better results. Try adding details like the colors, elements, lighting, or camera view.
2. Try Different Art Styles
Referencing specific art styles can be handy while creating an AI image. you may name renowned artists (or artworks) and different styles, like abstract, cubism, still life, or portraiture.
3. Use Image Prompts
Instead of trying to describe your idea, you’re able to upload an image and ask Bard to generate a similar one with the desired modifications.
For example, we uploaded a picture of a dog with a stick in his mouth and asked Bard to recreate it in the style of Vincent van Gogh. you’re able to see the results above.
Ensure you comply with copyright restrictions when using image prompts to generate AI images.
4. Keep Prompting
It’s unlikely that you’ll get the exact image you imagined on your first try. So, the key is to keep prompting and instructing Bard about specific modifications, like making the image realistic, using more vibrant colors, and so on.
Bard offers a simple and free way to generate AI images, so it’s worth giving it a go to see what results you may muster. You can even use Bard itself to generate prompts for AI images, or get inspiration from theseAI art prompt ideas.
