If you are a Google Bard user, you would know that the popular Gemini-powered chatbot lacked the ability to generate images like some of its counterparts from rival AI-centric companies such as OpenAI. However, the search giant has finally rolled out the ability for Bard to generate images for all users.
Now, when you open Google Bard by visiting bard.google.com, you will be greeted with a small prompt saying that you can generate images. “Begin your prompt with ‘Create an image of…’ or ‘Generate an image of…,’ and describe the picture you have in mind,” Google notes.
Once you have your image, you can tap ‘Generate more’ to get even more options. Once you generate your images, you can quickly tap the download button to get the image(s) offline.
We tried using the tool, and from our experience, it works as advertised. You can simply submit a prompt with key details such as the render style, characters, and even the lighting.
In our test, we asked Bard to generate a photorealistic image of a black hole, and the result was quite compelling. We took it a step further to ask if it can generate what is inside one, but the chatbot is smart enough to suggest that light can’t escape from a black hole; it is impossible to know what the inside looks like.
That said, Google Bard’s new ability works using the Imagen 2 text-to-image models and not Gemini Ultra at the moment. Also, generating images using Bard is absolutely free, unlike ChatGPT Plus, which is offered only as a paid subscription service. This is a significant advantage for Bard, as more people can use its image generator and spread the word.
Comments
0 comment