Day 6 of Free AI Course for Beginners: Google Gemini

Day 6 of Free AI Course for Beginners: Google Gemini

By now you must have learned a lot about AI, and we have already made five classes on AI. If you haven't watched them yet, please watch them first so that your base becomes very clear. And today, in Day 6, we will do a practical session on another AI tool that is quite interesting and quite popular — Google Gemini.

Welcome to Day 6 of the Free AI Course for Beginners. After ChatGPT, next we look at Google Gemini. You have definitely heard this name before. Google Gemini is a fairly popular AI tool, and we can do a lot with it. With its language model, we can do the same things we did with ChatGPT — we can ask it questions, get text written, get scripts written. So let's learn this practically too — how to use Google Gemini. You can use it on your phone as well as on your PC.

To get started, search on Google for "Google Gemini." You'll see the link gemini.google.com — open that link, and just like before, sign in and create your account. And exactly like ChatGPT, here too we can ask it things. It says "Ask Gemini" on the screen, so just like ChatGPT, we write a prompt here.

Example 1: Planning a Trip

First, let's give it a role. Imagine you're going somewhere for a trip — say, to Goa for three days — and you don't know where all to visit there. Gemini can help you with that. So the prompt is:

"Act as a professional travel planner. Make a three-day Goa travel plan. Also tell me the locations, vegetarian food options with the best places, and the best markets — because I also want to shop well — so tell me the best markets, timings, and vegetarian food options."

So we gave it a role, and then told it exactly what we needed. There's a Submit (Generate) button — click on it, and see how beautifully it creates an itinerary for you. It came back with a "Three Day Vegetarian Goa Delight Itinerary," broken down by Day 1, Day 2, Day 3 — really nicely laid out and good-looking too.

For example: at 8:00 AM, have breakfast — where? At Navtara Veg Restaurant in North Goa. Then at 9:30 AM head to the beach — Baga Beach is suggested. At 12:00 PM head out to a fort. It even gives timings along the way. At 3:00 PM, head out to watch the sunset at Chapora Fort — the same fort from the movie Dil Chahta Hai. And it also tells you where to have dinner at 8:30 PM. Then it gives the Day 2 plan, followed by the Day 3 plan. And at the end, it even compiled a complete list of vegetarian food options — which places have good vegetarian food, and what each place is known for. For example, Little Italy is known for pizza and pasta, and a place called "The Rasoda" serves Maharaja Thali.

You can use Google Gemini the same way — give it prompts, and it will help you out.

Nano Banana: AI Image Generation

Now, within this same Gemini, we have our next AI tool: Nano Banana. This is basically an AI model for generating images. Inside Google Gemini itself, we can create very professional-looking AI photos.

Earlier, we wrote text and got a text answer back. But now we'll write text and get an image back. There's a banner at the top saying "Nano Banana Pro is now on Gemini 3 Pro" — this is what lets us generate images. There's a banana icon along with a "Create Image" option — click on that, and just like before, write a prompt describing what kind of image you want.

Example: An Indian Housing Society

For example, say I want a scene like a Delhi housing society, where people live in flats and such — and you can't find such a photo on Google. What do you do? You get AI to create it for you, and it's very easy. You need to be direct and to the point here — you don't necessarily need to give it a role like "you are a photographer." You just need to add the details you want to see in the photo. Focus on the specific details you need.

For instance, I want to create that housing society image — a cheerful Indian society, Delhi-style — colorful, with people in it. So the prompt:

"Create a cheerful Indian society in New Delhi."

I even added a reference for style — something like Gokuldham Society from the show Taarak Mehta Ka Ooltah Chashmah. I mentioned who should be present — families, aunties, uncles, and of course, kids ("Tapu Sena"!). I described them as standing and chatting: "standing and chatting in the society compound with buildings, scooters and plants around." And for the overall mood: "bright, colorful and happy atmosphere."

I clicked Submit/Generate — and look, it created the housing society scene. Such an accurate photo came out!

And it's not just about recreating existing things — you can create things that don't even exist. Like a monkey wearing a coat and pants, eating ice cream on the street — literally anything you can imagine, you can create.

Example: A Crazy Prompt

Let's try something quirky. I wrote:

"A monkey wearing a torn kurta pajama, selling tea in the old lanes of Delhi from an old cart, with people coming to drink tea near him."

Just a simple prompt like that.

Changing the Image Format

Now, notice the image we generated is in a square shape. If you want a YouTube-style format instead, you can create that too. If you want a phone/reel format (tall/vertical), you can do that too. For that, just add it at the end of your prompt.

For example, for horizontal/YouTube format, you say 16:9 ratio. For vertical/phone format, you'd say 9:16. So I added "16:9 ratio" (with colons in the middle) at the end of my prompt and hit Submit.

Choosing a Style: Realistic vs. Cartoon vs. 3D

If you want a specific style — cartoon style, or completely realistic — you can mention that right at the start of your prompt. For example, I started with:

"Create a realistic scene of..."

I clicked Submit, and it created a very realistic image — the monkey in torn old clothes, selling tea, with people around drinking tea near him. It really looked real.

Then I copied the same prompt (there's a copy icon next to each prompt for reuse) and pasted it, this time replacing "realistic" with "3D Pixar style." So the prompt became:

"Create a 3D Pixar style scene..."

And look — it created a perfect scene in 3D animation style. Really impressive work.

Paid vs. Free Plan Differences

One important note: if you're using the paid plan of Google Gemini, the 16:9 ratio instruction works much more reliably. It might not work properly on the free plan — when I tried it on free, it kept generating everything in square shape regardless. So if you have a paid plan (note: with a Jio SIM you get a one-year free plan included), this feature works better.

Also, on the paid plan, you get access to an upgraded "Thinking" model — a newer version called "Think with Gemini 3 Pro" — which is only available on the paid plan. On the free plan, you only get the "Fast" option.

I tested the same prompt with the free/Fast option, and it messed up the 16:9 ratio request completely because it's a lighter model. Note also that on the free plan, you can only generate a limited number of images per day — maybe three or four. But overall, the image quality was still good, even if the aspect ratio wasn't respected.

Editing Photos with Nano Banana

You can even edit your own photos using Nano Banana. Earlier, you might have used Photoshop or PicsArt for photo editing — but now, photo editing is possible right inside Nano Banana. And you don't have to sit and manually edit anything — you just write a prompt, and the AI does the editing for you.

Step-by-Step Editing Example

1. Click the "+" (Add Files) button and upload your own photo.
2. Suppose you're wearing a green sweatshirt in the photo, and you want to change it. Just write: "Make me wear a denim jacket and white T-shirt."
3. Submit — and see, it kept everything else exactly the same: my face, the background — only the clothes changed to a denim jacket and white T-shirt, exactly as requested.

Changing the Background

Suppose now you want to change the background too — say, to a luxury five-star hotel room. Just write:

"Change the background to a luxury five-star hotel room."

Submit — and it changes the background. You can keep giving follow-up prompts in the same chat to keep refining the image. If you want to start something completely new, use the "New Chat" option on the side; if you want to keep editing the same image, just continue in the same chat below.

Fixing Lighting

It changed the background, but the lighting didn't quite match at first. So I added:

"Match the room lighting with me so that it should look realistic and not edited."

Submit — and surprisingly, it matched the lighting perfectly! The room had a warm tone, and it applied that same warm tone to me as well. Really impressive.

Adding Blur Effect

If you want a blur effect on the background (like a DSLR photo), you can ask for that too:

"Add blur effect to the background so that it looks like the image is captured with a DSLR camera using 50mm."

And sure enough, it added a nice blur to the background — before, the background was in sharp focus; now it has a nice bokeh blur, just like a real DSLR shot. You really can do great editing using Nano Banana.

Wrapping Up

Now you know how to use the Nano Banana AI tool — definitely give it a try. So today you learned about Google Gemini — its language model and its Nano Banana image model — and hopefully everything made sense to you easily.

How did you like this class/video? Let us know in the comments below — your comment, your like, and your share help us a lot. So please do support us

ChatGPT Full Tutorial - Master It in Just 10 Minutes!

What Is Prompt Engineering? How to Actually Talk to AI (Free AI Course – Day 4)

Machine Learning & Algorithms Explained Simply