Make it Pop #05 - Nano Banana Pro Special Edition

I’ve spent the last three weeks stress-testing Nano Banana Pro (Gemini 3 Pro Image).

Through my work, I had early access to the model before the official launch, so I’ve had time to really push it and see what makes it tick. I’ve also been meticulously analyzing the flood of content and stress tests from the community since the public drop.

If you look closely at the specs, you’ll realize something fundamental has changed.

The era of complex prompt engineering is over with this model. You no longer need complex and long prompts to get to the desired output.

We are entering the era of context engineering. Sounds complex, but it’s very simple.

Old models are blind. You had to describe a scene to them word-for-word. This new model can see, read, and reason. It doesn't need your adjectives; it needs data input.

RISE is a mental model you can use to master Nano Banana Pro.

The goal is to stop describing the output and start anchoring the input.

• R - Reference (the anchor): Stop describing your brand. Upload your "Source of Truth" (PDF Guidelines).

• I - Insight (the brain): Stop asking for "cool pictures." Ask for market context. "Search for the current trends in X and visualize them."

• S - Specification (the medium): Be technical. "Isometric 3D," "Vector Line Art"

• E - Execution (the lock): Enforce the rules. "Strictly use the uploaded font and colors"

3 "ANCHOR" WORKFLOWS TO TRY

Here is how applying that strategy solves business problems.

1. Web design

Problem: Most AI-generated apps look exactly the same. They are "vibe-coded." It’s the visual equivalent of seeing "—" (em dashes) in a text; you instantly know AI made it because it lacks structural intent.

Nano Banana Pro way: Use the model to generate code + UI images, which you can then input into vibe coding tools.

Prompt: "Based on this brand guideline and hero product - cardigan, create a high-fidelity website landing page for this brand ‘Khulan Cashmere’. I want something sleek and modern luxury"

2. The brand guidelines enforcer - Marketing

Problem: You have a beautiful Brand Guideline PDF, but AI tools don’t accept pdf input. You spend hours fixing the wrong shade of blue in Photoshop.

Nano Banana Pro way: You don't need to describe your style; you just need to upload the manual as pdf or image formats.

The Prompt: Based on this brand guideline and images, create an Instagram ad for our new traditional Mongolian cashmere cardigan

3. "Live" context render - strategy & creative direction

Problem: AI models are usually "frozen in time." If you ask for a street style shot, it gives you fashion from 2023. It doesn't know what is happening now.

Nano Banana way: This model can connect to the live internet.

Prompt: "Design a realistic pop-up store concept for our brand in Regents Street, London. Step 1: Search for the current weather in London right now. Step 2: Analyze the brand guidelines and hero product attached. There is a real yurt inside made of cashmere."

NANAO BANANA PRO

Technical specs

• Resolution: 1K, 2K, 4K

• Aspect ratio: 1:1, 3:2, 2:3, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9

• Input: pdf, url, text, image formats, code

• Output: text, png, jpeg, code

• Max input per prompt: 14 images, 49k words (61 pages pdf if ~800 words per page)

• Grounding: Connects to Google Search, World Knowledge

• Thinking: Model summarizes its reasoning process for outputs

• Watermarking: SynthID watermarking embedded

Top Features

• Enhanced Realism: Captures intricate textures and lifelike details.

• Text Rendering: State-of-the-art, multilingual (10+ languages)

• Connects to Google Search: for up-to-date facts

• Studio-quality controls & editing: camera focus, type, lighting, and more

• Advanced compositing: Up to 14 images for great quality

• Consistency: Character & Style preservation across shots

• Thinking Model summarizes its reasoning process for outputs (not all platforms offer this; if you use Google AI studio, you will see)

• Multiple outputs per 1 prompt 4 outputs + ask for more in the same convo, though currently limited/disabled on some platforms

I created a guide with showing the top feature examples

Key Takeaway

You don’t need any more complex and long prompts; the model has excellent reasoning and can make sense of your intentions even with simple prompts. Be a Context Engineer - it’s a mental shift. Feed the model your documents, live data, and your constraints, and let the reasoning engine do the rest.

• Google announced Nano Banana Pro, an upgraded version of its image model with sharper detail, better realism, improved character consistency, and higher-resolution outputs.

• Black Forest Labs released FLUX.2, a new image-generation model with sharper detail, stronger composition, better text rendering, and major improvements to realism and artistic control.

• Meta released new SAM models that can detect objects in images and instantly turn them into 3D reconstructions

• Hunyuan announced a new upgrade to its open source video-generation model, boosting motion quality, clarity, and prompt control while expanding global access.

• Warner Music Group has partnered with Suno to develop new AI-powered music tools and experiences. This could let artists create and release AI-enhanced music with label support, bringing AI from the sidelines into the mainstream music industry.

• Midjourney now lets you build a customizable public profile (username, bio, banner, profile pic/video) to showcase your AI art and follow other creators. It gives creators a personal, shareable portfolio, making it easier to present your style, connect with peers.

🏆 AI creator competitions worth joining

If you’ve got a video or concept brewing, these competitions are open right now, and they’re giving real prizes + visibility to your creative AI work

On a personal note, I have been super busy with the Gemini 3 and Nano Banana Pro launch at work, so I haven't been posting much on LinkedIn. It was intense, but I am so proud to be part of the launch team! It was such an amazing experience working with fantastic people and seeing the feedback from folks.

I also traveled to San Francisco to speak at a startup event on UX for creative applications.

After the two launches & event, I went to Aruba to celebrate my birthday and do lots of water sports. Sea bobbing, scuba diving, kite surfing, and just diving deep. Moi and the ocean became one lol, Aquaman vibes.

Hope you had an amazing Thanksgiving for those who celebrate!

Until next week,

Khulan