I created these images with just the voice and the new Gemini feature. I am impressed with the results

Google has just implemented the Image 3 image creation engine in Gemini, opening the doors for any user of this artificial intelligence to create images. This AI can be used on the web, but its potential on mobile seems pretty wild to me. In my case, I have Gemini as a replacement for Google Assistant, so I just have to press the power button to activate it.

By opening Gemini directly, I can give it instructions with my voice, and I don’t need to type the prompt. Hitting a button and telling it to “generate an image of a Mercedes SLR McLaren on a speedway” and it creates it perfectly is something I didn’t expect.

Image 3 is the new version of the artificial intelligence model that, from now on, Gemini will use to generate images. It is the most recent version and, according to Google, it has improved these three pillars compared to previous versions:

• The images have much more detail and fewer imperfections.
• The images are much more realistic than before.
• It is now possible to generate images with many more styles.

Using Image 3 in Gemini is more than simple. You just have to download the Gemini app from the Play Storeand generate the prompt. It can be written, but doing it with voice is much more comfortable.

As a good car lover, I have asked Gemini to generate an image of a Mercedes SLR McLaren on a speed circuit. I have chosen this model because it is quite particular in terms of design and, to my surprise, it has recreated it -almost- perfectly. You can still tell that it is an image generated by AI, but for a photo created in five seconds with my voice, it’s not that bad.

In the second image, I created something simpler, a dog and a cat walking together. The bokeh it has created is simply spectacular and, in terms of sharpness, the image is brutal.

To complicate things a little and see how you interpret the styles, I asked you for a street of a New York street, with backlights and CyberPunk style. The result, again, is surprising.

As an end to the holidays, I asked you for an image of an iMac, on a desktop, and on whose wallpaper the word ‘Xataka’ appears. Text generation is a problem for many free AIs. This is not the case with Image 3. It also has no problems generating brand logos, although it is not as fine here.

Image | Xataka

In Xataka | 19 pages and services to create images from scratch using artificial intelligence