Mishaal Rahman / Android Authority
As we approach the end of the year, the companies at the forefront of AI are rushing to get their latest and greatest models into the hands of consumers, developers, and businesses. Last week, OpenAI unveiled ChatGPT 5.1, a smarter model with more personality. Not to be outdone by its rival, Google has unveiled Gemini 3, the latest version of its AI model that the company says has improved reasoning, coding, and multimodal capabilities.
I spent the weekend testing Gemini 3 Pro, and I found these claims to be mostly true. Gemini 3 Pro accurately analyzed complex charts I gave it, threw together easy-to-understand slideshows from dry text documents, and even created bespoke Android apps based on a short description of my requirements.
Gemini 3 also introduces a small but incredibly useful new quality-of-life feature: the ability to proactively recommend follow-up questions. You might think this is useless, but that’s only if you’re researching a topic you already know. If you’re learning about something new, Gemini’s recommended follow-up questions serve as a great jumping-off point to drill deeper into complex subjects.
During my brief time with Gemini 3 Pro, I challenged it to answer 9 complex reasoning and coding questions. Here’s how it went.
Mishaal Rahman / Android Authority
Question #1: How Gemini 3 optimized my workout regimen I’ve recently started taking my health more seriously, and as a result, I’m regularly hitting the gym. My younger brother, who started his own fitness journey a few months ago, helped me build a full-body workout regimen consisting of the following exercises: Iso-Lateral Row
Chest Press
Shoulder Press
Lat Pulldown
... continue reading