AI How-To Series · Part 1
AI image tools are finally catching up with how creatives actually work.
Google DeepMind’s new Nano Banana Pro (inside Gemini 3 Pro) is one of the first models that feels less like a toy and more like a visual instrument: it can fix text inside images, shift lighting, change angles, upscale to 4K, and even turn a messy sketch into a production-ready render.
In this first “how to” guide, let’s walk through practical ways to use Nano Banana Pro so your visuals look like they came out of a studio, not a prompt test.
Step 1. Get into Nano Banana Pro
You don’t need a separate app.
You can access Nano Banana Pro directly from the Gemini ecosystem:
Gemini on web or app – the fastest way to try image creation and editing.
Google AI Studio – for more control over prompts and outputs.
Gemini API / Vertex AI Studio – if you’re building this into internal tools or workflows.
For this guide, assume you’re in Gemini on the web:
Go to Gemini.
Click “Create image” under the text field – that’s where Nano Banana Pro lives.
Step 2. Turn Rough Scribbles into Polished Visuals
This is where it stops feeling like “AI” and starts feeling like magic.
Use case: you have a rough logo idea, product shape, or layout sketched on paper or an iPad.
How to work:
- Upload your sketch to the Nano Banana Pro interface.
- Add a clear, simple prompt, e.g.
- “Turn this into a clean logo for a youth streetwear brand, bold and playful, flat color, no gradients.”
- Let Nano Banana Pro reinterpret the sketch into a finished-looking visual, logo, product render, or even a 3D-style building.
Think of the sketch as direction, not a final layout. The model is very good at being “inspired by” rather than copying every line.
Step 3. Generate a High-Quality Base Image
Even if your end goal is editing, start with a strong base.
Nano Banana Pro can generate super high-resolution, true-to-life images from a single prompt, anything from cinematic scenes to product hero shots.
Tips for stronger bases:
- Be specific about subject, setting, mood, and detail level.
- Mention camera framing (close-up, mid-shot, wide shot) if you care about composition.
- Add where it’s meant to live: “for e-commerce PDP,” “for OOH,” “for Instagram Reels cover,” etc.
This first generation becomes your “master visual” for later edits.
Step 4. Play with Colour and Lighting (Without Re-Lighting the Studio)
Traditional retouching teams spend ages nudging light and colour. Nano Banana Pro does this via a prompt.
You can:
- Turn day into night while keeping the subject consistent.
- Add more dramatic contrast and directional light.
- Create stylised lighting moods (soft morning light, neon city, chiaroscuro, etc.).
Workflow:
- Generate or upload an image.
- Prompt around the same image, e.g.
- “Make this a nighttime scene, keep the character identical, but light up the city with warm window lights.”
- Iterate until the lighting feels both dramatic and believable.
This is especially powerful when you need variations of the same visual for different seasons, campaigns, or price points.
Step 5. Explore New Angles and Shot Types
You don’t always get the perfect angle in the first render. Nano Banana Pro lets you “re-shoot” via prompt.
You can ask for:
- A wide establishing shot for banners.
- A tight product close-up for PDP.
- A different angle on the same moment or environment.
Think like a director:
“Show this scene from a low angle, looking up at the product, with the background softly blurred.”
This is powerful for storyboards, campaigns, or eCom: one idea, multiple framings, without redoing the entire shoot.
Step 6. Upscale and Reformat for Every Platform
Once you’re happy with a visual, Nano Banana Pro can upscale and resize without destroying quality.
- Upscale to 1K, 2K, or 4K, depending on where it will run.
- Change aspect ratios 16:9 for video, 1:1 for feeds, 9:16 for vertical formats.
Instead of re-creating for each format, you can prompt:
“Upscale this to 4K and adapt it for a 9:16 vertical hero image, keeping the composition centred on the product.”
For brands and agencies, this is where you start to see serious production savings, especially when multiplied across dozens of SKUs or campaign lines.
Step 7. Fix and Localise Text Inside Images
One of the most painful parts of AI images has been broken or nonsensical text.
Nano Banana Pro specifically tackles this. It can read and accurately replace text inside an image, for example, turning all English labels on a product into Korean, while leaving everything else intact.
That unlocks:
- Fast market localisation of packs and key visuals.
- Regional campaigns without restarting the creative from scratch.
- Cleaner typography for ads and social.
Prompt along the lines of:
“Translate only the visible English text on these cans into Korean. Keep all colours, fonts, and layout identical.”
For global brands, this is a quiet revolution.
Step 8. Generate a Whole Image Set in One Go
Instead of one hero visual per prompt, Nano Banana Pro can produce a sequence or set of images from a single, carefully written instruction: a story told across multiple frames, in one consistent style.
This is ideal for:
- Storyboards
- Comic-style explainers
- Social carousels
- Variant testing on performance media
Think: one detailed prompt → ten coherent frames you can curate, refine, and version.
What This Means for Creative Teams
Tools like Nano Banana Pro are changing the centre of gravity in production. You no longer need a full shoot every time you want to explore a new idea, test a variant, or localise a visual. You need clear creative direction, strong prompts, and a system that keeps everything on-brand.
That’s exactly where HubStudio comes in.
We’re a creative-led AI studio: we combine AIGC engineers, senior creatives, and real studio capabilities to design AI that actually understands your brand DNA, then scale it into thousands of images, short videos, and eCom assets, fast, on budget, and on brand.
