Google is constantly working on its various AI models and recently added a new function to Google Gemini 2.0. If you use this service on your computer in Google AI Studio, you can now not only generate images but also edit them. Does Photoshop need to look over their shoulders now? Perhaps! We’ll show what you need to do to use the function for free with unlimited tries.
Yes, I know there are many AI tools to generate images. However, some of them come with a hefty price tag, while others are very limited and only allow you to generate a few images per day. Hence, it’s great Google has launched an experimental version of Gemini 2.0 Flash as part of its Google AI Studio. It lets you create images free of charge and virtually without any limit (“only” 1,000 edits per day). Something tells me it is going to be more restrictive the more popular it gets.
Google AI Studio: What you need to do to create and edit images
If you are not yet familiar with Google AI Studio: Google AI Studio is a web-based tool from Google that allows you to work with Google’s large language models (LLMs) such as Gemini. Think of it as a toolbox where you can experiment as much as you like. If you were to click Create Prompt on the left, you will see an interface on your computer that is easy to navigate. You have a bar there which you can enter your prompt. Sounds familiar so far?
The Studio has many more options, but for today, we only want to concentrate on image editing. To do this, it is important you use the correct Gemini version within this Studio. If you were to go to the normal Gemini site, it won’t work. This makes it a little tricky if you don’t know how to find your way around, but that’s exactly what I’m going to tell you step-by-step:
- In your computer browser (it doesn’t work on mobile yet), launch the Google AI Studio.
- Now select the correct model, Gemini 2.0 Flash Experimental, that is located on the right-hand side.
- Directly below, under Output Format, make sure that Images and text is selected, otherwise you will not be able to edit any images.
- At the bottom of the input line, you can now formulate a prompt for an image.
- Alternatively, you can click on the plus symbol in the input line on the right and then upload your own images.
- From here, all you need to do is use your creativity to edit an existing image. Just give it a try. You can change people’s poses, dress them differently, give the pictures different backgrounds and so much more.
What else can image editing in Google AI Studio do?
If you are on the empty prompt page in Google AI Studio, you will see three examples of editing options in the middle. A distinction is made between “Image Editing”, “Visual Story” and “Birthday Card”. If you were to click on any of these three examples, you would get a very clear demonstration of what the tool can do. Logically, you can edit an image under “Image Editing”. The example shows croissants that have been given a drizzle of chocolate by the prompt.
However, the other examples were also equally impressive. You can use the prompt to come up with a story, which is then accompanied by suitable images. The cool thing is, that the images remain consistent! This means that the cute sheep from the sample story looks the same in all the pictures. In addition, a birthday card demonstrates that Google could even compete with tools like Canva in this department.
However, this image editing mode could pose an even greater threat to programs like Adobe Photoshop. I’m not a big Photoshop fan, but you certainly need a bit of time and skill to let a person in a t-shirt put on a leather jacket. As you can see in the picture above, I began simply enough and replaced my grey shirt with other bright colors. Following that, I put on hats, slipped into an astronaut suit, and even dressed up as a pirate.
All of this can be done with short prompts, and once again, it’s important to keep the template consistent. Incidentally, this doesn’t always work, but Google is excused here. After all, it’s still the experimental version of the tool.
However, you can still get an early impression of how powerful the function already is by trying it out. For instance, I also redecorated my living room. I replaced the original pictures with maritime scenes before asking for a leather couch. Obviously, this genie in the bottle answered my wish and got me one, as you can see in the following pictures:
What does that mean for you? Perhaps the more important question is what does it mean for programs like Photoshop and Canva. A lot of things that you would previously have spent a lot of time on are now possible in the blink of an eye. Whether you’re designing party invitations, peppering a recipe blog with illustrated recipes, or filling a store with different layouts, it’s all done in a flash with Gemini 2.0 Flash.
For those in the professional sector, I imagine it would be exciting if you could wear different shirts virtually in many situations or completely rearrange and redecorate your home in double quick time. The possibilities are almost unlimited.
To be fair, however, it should be mentioned that you will notice this is still an experimental, early version of the tool. Previously requested accessories might reappear in later pictures, and sometimes, the scene does not remain consistent. Through trial and error, you should be able to achieve the desired result.
I always find it amazing what is already possible and for free! In any case, we hope you enjoy playing around with image editing. Feel free to tell us in the comments how well it works for you and what other image AI editors you use.
