Google announced Gemini 4 Argon in September 2026, a frontier model designed for complex cybersecurity workloads and built with a 1-million-token context window.
Key announcements
The release capped a busy month that also saw Gemini 3.8 Flash and 3.8 Flash Cyber. Google introduced expressive voice models via Gemini 3.8 Live, opened pre-orders for the Googlebook laptop, and launched the Gemini app for Windows. Scientific milestones included mapping human DNA in AlphaGenome Atlas, tracking global methane emissions from space, and deploying Project Suncatcher to test machine learning hardware in orbit. Fast Company named Google’s design teams the 2026 Design Company of the Year.
What it means for users
Developers and security teams can now access Gemini 4 Argon through the Fairwind Program. This rollout is limited to trusted cyber defenders. Google is gathering feedback to refine safety measures before making the model available to the wider public, enterprises, and consumers.
Gemini 3.8 Flash and 3.8 Flash Cyber offer improved reasoning and coding speed at the same low cost as the previous 3.7 Flash version. The Cyber variant provides autonomous vulnerability discovery and code patching for organisations enrolled in the Fairwind Program. These tools handle long-horizon software engineering and multi-step problem solving.
Users can converse naturally using Gemini 3.8 Live and 3.8 Live Extended Thinking. The standard Live model handles everyday audio requests. The Extended Thinking version manages multi-step reasoning without adding lag. Both models are available in Google AI Studio and the Gemini API. They also run across Search Live, Gemini Live, and Google Workspace applications like Gmail, Docs, and Keep.
Text-to-speech capabilities have expanded with Gemini 3.8 Flash-Lite TTS and Gemini 3.8 Flash TTS. These models allow the creation of custom voices and direct scene dialogue from simple text prompts. This replaces static presets with a dynamic studio for building voice-first applications in Gemini Notebook and Google Vids.
Connected Apps in Gemini now integrate productivity tools like Airtable, Linear, monday.com, and PandaDoc. Creative platforms including Adobe, Picsart, Squarespace, and Webflow are also linked. Lifestyle services such as Peloton, SeatGeek, and Experian are included. Users select these apps in their Gemini settings to run tasks without switching tabs.
Lyria 3.5 in Gemini generates richer music arrangements and coherent songs from text prompts or images. This feature is available globally in the Gemini app, Google AI Studio, Google Flow Music, and Google Vids.
Students can transform their study routine with new tools in Gemini Notebook. These include interactive coursework features and animated video summaries. Users can create custom flashcards, take quizzes, and view 60-second Short Video Overviews in over 80 languages. Eligible college students can claim a free 12-month Google AI student plan with higher usage limits.
Google Labs is expanding CC from a personal agent to a collaborative group agent. This version manages household logistics and shared schedules. CC receives its own email address to coordinate family calendars and track group to-dos. The feature includes built-in privacy controls and is rolling out as an experimental preview.
Google tools currently support more than 300 languages spoken by 7 billion people, reaching 86% of the global population. Open-sourcing models like MedGemma and partnering with research institutions helps scientists accelerate discoveries. New visualisations of the AI & Economy ATLAS data show how artificial intelligence is reshaping global work.
Project Suncatcher will launch its first test satellite with Planet. This mission evaluates Google TPUs under real space conditions. Satellites in low-Earth orbit can access up to eight times more solar power than on Earth, offering new possibilities for long-term AI infrastructure.
Pre-orders are open for Googlebook, a laptop built on Android with desktop foundations from ChromeOS. The device is engineered for Gemini Intelligence and pairs with an Android phone out of the box. New on-device features include Magic Pointer, which lets users ask Gemini about anything on the screen with a cursor wiggle. Rambler smart voice dictation structures verbal notes into organised text. No-code custom widgets and background task processing simplify daily workflows.
WeatherNext 3 is the most advanced global weather AI model. It includes real-time satellite data, hourly refreshes, higher resolution, precise precipitation forecasting, and clean energy variables. The model is integrated across Search, Gemini, Maps, Google Maps Platform, and Cloud.
AlphaGenome Atlas is a high-resolution map of human DNA. It predicts the effects of all 9 billion possible single nucleotide variants. Designed for users without coding skills, it accelerates breakthroughs in rare disease research and complex genetics.
Google partnered with NASA’s Jet Propulsion Laboratory to develop MAPL-EMIT. This deep learning model tracks global methane leaks from space. It detects 50% more emissions than human experts and has uncovered over 23,000 previously unmapped plumes. This includes 24 of the world’s 25 largest-emitting landfills. Data is openly available on Google Earth Engine.
Scientists at Google, HHMI Janelia, and Cambridge University built the first complete brain map of an adult male fruit fly. The research charts over 166,000 neurons and nearly 12,000 distinct cell types with Google AI. Stitching millions of microscopic images into a navigable 3D model helps understand how sensory cues transform into movement.
The DeepMind Institute serves as a platform for researchers and thinkers to publish ideas about a world with artificial general intelligence. The institute aims to address the risks of AGI and build a consensus that ensures benefits for everyone.
The Gemini app for Windows allows users to access help using the Alt + Space shortcut. This opens Gemini to fact-check documents or brainstorm ideas. In the dedicated workspace, users can delegate multi-step tasks to Gemini Spark, synthesise files across Gmail and Drive, and generate Nano Banana images or Gemini Omni videos.
