Skip to main content
Two hosts sit in a studio, microphones and laptop screens showing Google Gemini 3, ChatGPT and Nemotron 3 logos.

Editorial illustration for Google Launches Gemini 3 Flash, Sets as Default Model in Gemini App

Gemini 3 Flash: Google's Nimble AI Model Redefines Speed

LWiAI Podcast #229: Google defaults Gemini 3 Flash ChatGPT app launch Nemotron 3

Updated: 3 min read

Google is making Gemini 3 Flash the default. OpenAI opens an app store. And Nvidia steps into the big leagues with Nemotron 3.

This week on the Last Week in AI podcast, the team unpacks a firehose of headlines. A major model rollout collides with a fresh marketplace for ChatGPT plugins. A chip race heats up, both in Beijing and inside Amazon’s AGI group.

And yes, Anthropic’s mysterious $10 billion Broadcom deal finally gets a name. The episode runs fast, nearly 70 minutes of banter, analysis, and the occasional rabbit hole. From Google’s quiet model swap to China’s Manhattan Project for silicon, from OpenAI’s sudden equity overhaul to a vibe-coding startup valued at $6.6 billion, there’s no shortage of signal amid the noise.

Here’s the full breakdown, timestamp by timestamp.

Notable releases include OpenAI’s GPT-5.2 Codex for advanced coding and Google’s Gemini Free Flash for competitive AI application performance.

And so, we find ourselves in a week where Google quietly rewires its core product, OpenAI opens a storefront, and Nvidia becomes a model maker to rival the labs it once supplied. Each headline pulls in a different direction, consolidation here, fragmentation there, but the signal is the same: the infrastructure race is becoming the application race, and the application race is becoming the talent race. China’s Manhattan Project meets Amazon’s $10 billion bet; Anthropic gets its mystery check from Broadcom; a vibe-coding startup commands a $6.6 billion valuation.

The ecosystem is not just growing, it’s reordering itself. Meanwhile, researchers ask whether AI could suddenly accelerate in 2027, as if the present pace weren’t enough to keep us awake. We don’t have the answer, but we have the timestamps, the paper links, and the conviction that the next episode will demand a new set of questions.

Until then, keep reading, keep building, and keep an eye on what defaults.

Common Questions Answered

What makes Gemini 3 Flash different from previous Google AI models?

Gemini 3 Flash is a lightweight model designed for faster and more responsive interactions. It represents Google's strategic approach to delivering more efficient AI performance, with a focus on real-time responsiveness and user experience.

Why did Google choose to make Gemini 3 Flash the default model in the Gemini app?

By setting Gemini 3 Flash as the default model, Google is signaling a commitment to improving AI accessibility and immediacy for everyday users. The move suggests a strategic push to make AI interactions more approachable and quick, prioritizing speed and efficiency.

How does Gemini 3 Flash fit into Google's broader AI strategy?

Gemini 3 Flash appears to be part of Google's effort to simplify AI interactions and make the technology more user-friendly. The lightweight model indicates Google is focusing on delivering quicker AI responses while maintaining core functionality, positioning themselves competitively in the rapidly evolving AI landscape.

LIVE03:21OpenAI's Miles Wang in Talks for USD 2B AI Drug Discovery Startup