In partnership with
Do you switch AI models when new ones launch?
Gemini 3 Pro just went live.
The anticipation building up with the leaks and rumors over the past week was sky high. Polymarket odds hitting 92%. It was only a matter of when, not if.
That changes today. I managed to get early access and it does NOT disappoint.
On paper, the numbers look absurd:
91.9% on GPQA Diamond [vs GPT 5.1 is below 90%]
76.2% on SWE-bench Verified [vs Sonnet 4.5 is now #2 at 70.6%]
81% on MMMU-Pro
#1 on LMSYS Arena at 1501 ELO. [vs Grok 4.1 just released yesterday at 1483]
Literally the #1 in everything now.
However, I gave it a spin myself to stretch its imagination and see how far it gets me. So I gave it three tests and here's how it went.
Test 1: Generate a 3D model of graphene
Prompt: "Generate a simulation of the molecular structure of graphite."
Ten seconds later: fully rotatable 3D molecular structure on screen. You can zoom, rotate with your mouse, inspect individual carbon atoms bonded to three neighbors in the honeycomb pattern.
Gemini generated working code that renders the structure with physics and interaction built in. The molecular structure is accurate.
Test 2: Create a game
Prompt: "Create an office simulation game using three.js."
Gemini made an office simulator where you control a character moving around. The twist: you only get paid when you're moving. Stand still at your desk and your salary stops.
I didn't ask for that mechanic. Gemini decided that's what makes it fun.
Isometric view, movement controls, collision detection, salary counter updating in real-time. Total time from prompt to playable: 15 seconds.
Test 3: Make a presentation about itself
Prompt: "Create a presentation making the case for Gemini 3 Pro's capabilities."
Seven slides. Title, problem statement, features, benchmarks, use cases, closing.
Standard pitch deck structure: opens with why AI reasoning matters, builds through evidence, closes with applications. Formatting is consistent throughout. Fonts match, information hierarchy works, visual balance across slides.
What's available today
Gemini 3 Pro: Live now in Gemini app (iOS/Android/web), Google AI Studio, Vertex AI, Search AI Mode. Free tier included.
Google Antigravity: New IDE where agents autonomously plan, code, test, and ship apps. Public preview at antigravity.google, completely free. We’ll cover this in an upcoming edition.
Generative UI experiments: Visual Layout and Dynamic View rolling out in Gemini app and Search AI Mode. My favorite update so far.
Gemini Agent: Gmail triage, calendar management, deep research. Google AI Ultra subscribers, US only.
Until next time,
Vaibhav 🤝
PS. One more thing. This isn't everything Google's shipping.
If you read till here, you might find this interesting
#AD 1
Shoppers are adding to cart for the holidays
Over the next year, Roku predicts that 100% of the streaming audience will see ads. For growth marketers in 2026, CTV will remain an important “safe space” as AI creates widespread disruption in the search and social channels. Plus, easier access to self-serve CTV ad buying tools and targeting options will lead to a surge in locally-targeted streaming campaigns.
Read our guide to find out why growth marketers should make sure CTV is part of their 2026 media mix.
#AD 2
The Simplest Way To Create and Launch AI Agents
Imagine if ChatGPT, Zapier, and Webflow all had a baby. That's Lindy.
With Lindy, you can build AI agents and apps in minutes simply by describing what you want in plain English.
From inbound lead qualification to AI-powered customer support and full-blown apps, Lindy has hundreds of agents that are ready to work for you 24/7/365.
Stop doing repetitive tasks manually. Let Lindy automate workflows, save time, and grow your business.
