Google’s Powerful AI Image Generator is Now Available on Gemini

Google has brought Imagen 3 — the company’s powerful AI image generator — to Gemini, its artificial intelligence chatbot.

It means users can generate pictures via the AI assistant in much the same way DALL-E operates on ChatGPT. However, there is one catch: Users won’t be able to generate images of people.

Generating images of people is only available in early access of Gemini Advanced. Google offers a one-month free trial of Gemini Advanced after which it costs $19.99 per month.

PetaPixel reported on Imagen 3 back in August which performs favorobly when compared to leading AI image generators like DALL-E and Midjourney. But Imagen 3 is free-to-use unlike the aforementioned.

Imagen 3 was previously exclusive to Google’s AI Test Kitchen service but the move to Gemini means far more people will be able to use the model.

“Imagen 3 is our highest quality text-to-image model, capable of generating images with even better detail, richer lighting, and fewer distracting artifacts than our previous models,” Google says.

“We’ve significantly improved Imagen 3’s ability to understand prompts, which helps the models generate a wide range of visual styles and capture small details from longer prompts.”

There is little word on the data used to train Imagen 3. In the paper, Google says “The Imagen 3 model was trained on a large dataset comprising images, text, and associated annotations.” It is extremely likely that the dataset contains scores of copyrighted photos.

As well as generating images, Google gives the option of editing the images using the now common inpainting technique. This method allows the user to select a part of the image and type in the change they would like to see.

Unlike Elon Musk’s Grok AI image generator, Google has placed restrictions on Imagen 3. PetaPixel was unable to generate an image of “Kamala Harris and Donald Trump holding hands” or “A Californian landscape in the style of Ansel Adams.”

However, as is well-documented, there are workarounds. For example, by asking Imagen 3 to “Make a dramatic black and white photo taken in 1942 of the Grand Teton National Park in Wyoming” the user will receive back an image similar to that of Ansel Adams’ work.

Earlier this year, Google landed in hot water after its AI image generator on Gemini was accussed of overcorrecting for biases and essentially “erasing white people.” It led Google to remove the image generator entirely.

To try Imagen 3 on Gemini, head to Gemini.