Now we'll use the same Gemini Tool to generate and use images. For most, you'll click on the tools icon to select "Create Images"
Note: You'll want to be logged in to your Gmail account to use Gemini.
Use the "Fast" model - choose it from the "v" arrow in the lower right of the "Ask Gemini" box
Also be sure to choose the "Create Images" tool
Give the following prompts in sequence, waiting for Gemini to show you the results.
"Create a happy cat"
"Give it 5 kittens and a bowl of milk"
"Change the scene to nightime in a tropical forest"
When you're finished, get ready for a new chat by clicking on the word "Gemini" in the upper left.
This is trickier.
The idea is to use an image that already exists (a photo, or an image from a website, or a screenshot). To keep things as simple as possible I'm going to describe copying an image from a website. You could equally well use a photo you've taken that's on your computer or one that's in your Google Photos (if you use it).
Open a new web page and find one with images on it. Wikipedia.com is good.
Copy the image. ( Right-Click (Mac or PC), or long-press on an iPad)
Switch to your Gemini Page
Paste the image into the prompt box
Describe what you want to do with the image
"Remove the background" is a common command
Follow up as you did with the cat with whatever changes you want to make to modify the image.
When you're finished, get ready for a new chat by clicking on the word "Gemini" in the upper left.
The goal here is to take advantage of what the computer world calls "Multimodality". That's geek-speak for 2025-era AI's ability to take input from text, images, and sound.
Video 6:04
What we're going to demonstrate is extremely useful when you're having problems with your computer and a strange message pops up on the screen. You can "capture" that message, paste it into Gemini, and ask the computer a question about it, like: "What should I do?"
You could also add an image already stored as a file by clicking on the "+" key in the "Ask Gemini" box, but what I'm doing here is often extremely useful.
Explaining this is a bit complicated because how images are captured varies across the types of computers you're likely to use. I'll give overall instructions and then the specific key combination for each computer type
Capture an image of a screen region to your clipboard
Specific instructions for each computer type are below
Paste that image into the "Ask Gemini" Box
Add a text prompt that asks Gemini about that image
You can also tell Gemini to modify that image
How to capture a screen region on each computer type
Windows: Win-Shift-S
Opens the "Snipping Tool", which gives you directions
Mac: Cmd-Ctrl-Shift-4
Changes the cursor so you can identify the region to capture
iPad: Power+VolUP (or Home on older)
Tap thumbnail, then drag
Chromebook: Ctrl-Shift-ShowWindows key
ShowWindows Key is rectangle with two vertical lines to its right - somewhat like []||
Start Fresh: Click "Gemini" in the top left.
The Prompt: Type all lines (or copy/paste) the following command and press Enter:
"I want to create a birthday card for my grandson, Leo, who loves dinosaurs.
Open a Canvas.
Generate an image of a friendly, realistic, T-Rex wearing a party hat.
Below the image, write a funny, 4-line birthday poem for an 8-year-old.
Add a big title at the top that says 'Happy Birthday Leo!'"
3. Make the Title text multicolored, with candles.
Refining the Image (Optional):
If the dinosaur looks too scary, type: "Change the image to be a cute cartoon dinosaur instead."