Gemini can now transform your questions and complex concepts into customizable interactive visualizations directly in your chat. Adjust variables, rotate 3D models, and explore data for a more immersive way to learn and explore in Gemini.

To try it, head to on web, and select “Pro” from the model picker, and ask Gemini to “show me” or “help me visualize” a complex concept so you can see for yourself. What’s the first prompt you’re going to visualize?

i just tried this: "show me a sunpath diagram for a south facing house"... quite impressed with the result !👍

Honestly, this is what education has been missing… less guessing, more seeing and doing.

This is how I learn best, by seeing and interacting. Feels like a big step for truly understanding complex ideas.

Big step for Gemini. This is not just better answers. It is AI starting to generate the right interface for the question in real time. Once chat can turn complex ideas into interactive visuals, the UX stops being chat only and starts becoming adaptive software.

This is actually game-changing for learning! No more struggling with flat diagrams, ask Gemini to visualize quantum entanglement or blood flow in the heart and tweak it live. Finally feels like having a personal science lab in chat.

"Me in 2023: 'Gemini, please explain photosynthesis like I’m five.' Me in 2026: [Vigorously adjusting the UV index on a 3D interactive leaf while wearing a robe] 'BEHOLD, THE CALVIN CYCLE IN 4K, BRETHREN.'"

Great feature but prioritizing fundamental citation & UX problems needs to happen ASAP over features like this. The open web needs to see major improvements if Google wants to sustain a healthy ecosystem. Full default Frankenstein recipes need to go & our recipe domain page isn’t being cited a lot in AI outputs. Attached is a branded personalized full AI recipe output. Gemini knew that I was the creator of the popular brand Inspired Taste yet we aren’t cited at all in a full branded Frankenstein recipe riddled with errors. @rmstein and @rajanpatel, this is eroding trust in Google and our brand at alarming rates! cc @glenngabe @rustybrick @lilyraynyc

Holy

Guys I was wondering if you can also make law school studies easier. You know we can use some help over here too..like making charts and chain of thought visualization of some sort

:(

Google turning Gemini into a fancy PowerPoint with extra steps while their search is still hallucinating and their models lag on basic reasoning. Peak "features over fundamentals."

Really useful update, these updates make working on the Gemini app so much fun 🔥

the interesting bit here isnt the 3d models, its that chat interfaces are finally becoming dynamic. text in, simulation out. this is the end of "prompt then read a wall of text" and the start of actually exploring an idea. gemini shipped it first, everyone else will follow within a month

It's simply creating static images

@Google Back in my day, 'interactive learning' meant the teacher letting us use the colored chalk on the blackboard. Now these kids are asking AI to generate 3D models they can rotate in mid-air. We were out here fighting chalk dust, and they’re living in 2050! 😅

Legit answer as far as i know. Better than expected, great work, for real, wtf... i cant bealive this is on the tool i use daily!!!!!!!

i upload my recent paper and prompt 'help me visualize', but it failed

Finally, a chatbot that doesn't just tell me the answer but makes me see it. Now I can procrastinate productively by rotating 3D models instead of just staring at walls between builds

Yup was understanding token bucket today and it beautifully rendered this visualisation. This is a nice addition.

Any plans to make this available via API? I could def see this integrated in ! Very cool stuff for education!

now make it stop generating images eventhough we arent asking it to

Kinda seems like you guys are just a few weeks behind everything anthropic does

can’t wait to rotate a 3D model of a hallucination in real-time lol

Alright, let’s push this beyond just generating visuals. Try this: Create a short visual sequence (4 scenes) showing the invisible transformation of a human. from distraction to discipline to purpose to legacy. No text. No obvious symbols. Each scene should subtly evolve the environment, posture, and energy of the character to reflect internal change. Then explain your creative decisions. what changed, why it changed, and what it says about human growth. I’m curious to see if you can move from just creating images to actually understanding transformation over time.

game changer

Updates, updates, updates, and updates

cool feature but the real test is: does it still hallucinate the data it's visualizing? pretty graphs don't fix wrong answers 👀

@grok extract the list of all the prompts used in this video

@GeminiApp you know how to grab my attention, but could you for the love of GOD let Gemini search and reference past conversations within the app? optional mode or whatever, but it's getting ridiculous with the sheer size and number of threads I have by this point!

Dear teachers, Your jobs have been taken. Best regards, AI.

Yeah i got an image as well.. Not a good feature yet

visualize "how to talk to girls" and watch the entire server farm explode from the complexity 💀

This is so cool! Great work Gemini Team!

I like to study math on my free time, I take snips and upload to verify my answers are correct. chatgpt has limited uploads to help me verify my math is correct, the same for Gemini. I decided to check copilot and theirs is unlimited. Knowledge should be free.