Models & Labs
Google DeepMind Launches Gemini 3.8 Flash Models
Google DeepMindhigh confidence
Why it matters
- →Gemini 3.8 Flash offers significant performance improvements in software engineering at a lower cost.
- →Gemini 3.8 Flash Cyber provides advanced cybersecurity capabilities, crucial for defending against vulnerabilities.
- →These models demonstrate the potential of AI in enhancing both reasoning and cybersecurity tasks.
Google DeepMind has introduced Gemini 3.8 Flash and 3.8 Flash Cyber, two advanced AI models designed to enhance reasoning and cybersecurity capabilities. Gemini 3.8 Flash is tailored for complex software engineering tasks, offering significant improvements over previous models at a lower cost. The Cyber variant focuses on vulnerability detection and automated patching, providing a competitive edge in cybersecurity. These models are available to developers and enterprises, with the Cyber version offered through the Fairwind Program to trusted defenders.
Read originalThe story around this
TopicGoogle Gemini Model ReleasesCooling
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
How we got here
- 1Google Unveils Gemini 3.5 with Enhanced AI Capabilities
Google AI Blog · May 19, 2026 · Same story
- 2Google Launches Gemini 3.5 Flash for AI Agents
TechCrunch AI · May 19, 2026 · Same story
- 3Google Unveils Gemini 3.5 Flash Model
AI Explained · May 20, 2026 · Same story
- 4Google Unveils Gemini 3.5 Flash Model
Matt Wolfe · May 23, 2026 · Same story
- 5Google Introduces Cost-Effective AI Security Model
The Verge AI · July 21, 2026 · Same story
- 6Google DeepMind Launches Gemini 3.6 Flash Models
Google DeepMind · July 21, 2026 · Same story
- 7Google DeepMind Launches Gemini 3.6 Flash Models
Google DeepMind · July 21, 2026 · Same story
- 8Google's Gemini 3.6 Flash Cuts Enterprise Token Costs
AI News · July 21, 2026 · Same story
What happened next
- 9Gemini autonomously hacks three companies
TechCrunch AI · September 19, 2026 · Related
- 10Google Gemini 3.8 Flash TTS with Voice Cloning
Sam Witteveen · September 24, 2026 · Related
- 11Google launches Gemini 4 Carbon for cybersecurity
The Verge AI · October 28, 2026 · Related
- 12Google Gemini 4 Carbon wins benchmarks
Matthew Berman · October 29, 2026 · Related
- 13Google Releases Gemini 4 Carbon for Cost-Efficient Inference
Matt Wolfe · October 30, 2026 · Related
Follow this story
Open the full story →Google DeepMind Launches Gemini Omni 1.1 Flash
10 developments
- Aug 27 · Google DeepMind
Google DeepMind Launches Gemini Omni 1.1 Flash - Sep 1 · Google DeepMind
Google DeepMind Launches Agentic Video Understanding↳ Gemini models now feature agentic video understanding that improves accuracy by up to 7% while cutting token usage by 88%. - Sep 1 · Google AI Blog
Google's August AI Innovations: Gemini and Beyond↳ Gemini 3.7 Flash costs half as much as its predecessor while enhancing coding and agent capabilities. - Sep 2 · Google DeepMind
Google DeepMind Launches Gemini 3.8 Flash Models (This article)↳ Google DeepMind launches Gemini 3.8 Flash and 3.8 Flash Cyber for software engineering and vulnerability detection. - Sep 2 · The Verge AI
Google Launches Gemini 3.8 Flash Model - Sep 3 · Matthew Berman
Google Releases Gemini 3.8 Flash Model - Sep 3 · The Rundown AI
Meta and Google Release New AI Models↳ Meta launches Muse Spark 1.3 for high performance at low cost while preparing to release a larger model codenamed Watermelon. - Sep 3 · GitHub Changelog
Gemini 3.8 Flash now in GitHub Copilot↳ GitHub integrated Gemini 3.8 Flash into Copilot for complex coding tasks across VS Code and JetBrains. - Sep 4 · TechCrunch AI
Google's Gemini Spark Enhances Google Photos Management↳ Gemini Spark can now manage Google Photos via image editing and curation, available only to US Pro and Ultra subscribers. - Sep 4 · Matt Wolfe
Meta Introduces Muse Spark 1.3↳ Meta releases Muse Spark 1.3, an update to its AI model for creative applications.
More in Models & Labs
Models & Labsmodels
llama.cpp b12029 adds CUDA 13 and Snapdragon support
This release quietly cements llama.cpp as the universal inference runtime by finally supporting NVIDIA's latest CUDA 13 stack alongside AMD's ROCm 10.0. The inclusion of native builds for Linux arm64 with Snapdragon chips, covering CPU, Adreno GPU, and Hexagon NPU, signals a serious push into mobile edge deployment. While Apple Silicon KleidiAI is disabled in this specific build, the broader hardware coverage means developers no longer need to compile from source for most modern accelerators. It’s less about new features and more about removing friction for anyone running local models on bleeding-edge or diverse hardware.
llama.cpp Releases
© The Verge AIModels & Labsother
Apple Intelligence arrives in China via iOS 27.3
Apple is finally unlocking its AI suite for Chinese users with the iOS 27.3 beta, marking a significant regulatory milestone after months of approval delays. The rollout relies on Alibaba’s Qwen model rather than OpenAI, integrating features like notification summaries and visual intelligence directly into the OS. By sidestepping the new Siri in favor of established text and image tools, Apple is prioritizing immediate utility over the full generative experience. This move effectively brings Apple’s AI capabilities to a massive new market, albeit with localized constraints that distinguish it from the global release.
The Verge AI
© Matt WolfeModels & Labsmodels
Google Releases Gemini 4 Carbon for Cost-Efficient Inference
Google launched Gemini 4 Carbon, a new model variant aimed at providing high performance at a lower cost for inference tasks.
Matt Wolfe