Google just launched Gemini 3.5 Flash — here's all the upgrades
It looks like Google is preparing "Gemini Spark" to turn AI into an Operating System
Google’s vision for AI is undergoing a massive shift. Today, Google introduced Gemini 3.5 Flash as the latest in their family of models combining frontier intelligence with action. The faster and smart model delivers frontier performance for agents and coding, while excelling at complex long-horizon tasks that deliver real-world utility.
This launch marks a next-generation leap, one that's built entirely around “agentic workflows.” According to Google, the company wants AI to stop behaving like a chatbot you constantly have to manage, and start acting more like an invisible digital layer that can independently complete complex tasks from start to finish.
The next wave of AI at a glance
One thing Google has made clear: Gemini 3.5 is built for action rather than simply conversation. Instead of simply generating text or summarizing information, Gemini 3.5 Flash is designed to execute multi-step workflows, maintain software projects, prepare documents, coordinate “subagents” and handle longer-running tasks with far less human oversight.
According to Google, the model outperforms Gemini 3.1 Pro on several coding and agentic benchmarks, including Terminal-Bench 2.1 and MCP Atlas, while also dramatically reducing response latency. That's incredibly important because AI agents often feel frustratingly slow.
While tools from companies like OpenAI, Anthropic and Meta can already perform surprisingly advanced workflows, many still feel clunky when juggling multiple steps, browsing tools, or reasoning through complex tasks. Google promises that Gemini 3.5 Flash removes that tradeoff between reasoning power and responsiveness.
Google Spark proactively assists with workflows
But today, Gemini Spark may be the even bigger story. The persistent AI agent designed to quietly run in the background, takes actions on your behalf while remaining under your supervision.
According to Google, Spark can proactively assist with workflows, coordinate tasks and function more like an always-on digital layer than a standalone chatbot. Trusted testers will receive access first, with a broader rollout to Google AI Ultra subscribers in the U.S. expected next week.
Get instant access to breaking news, the hottest reviews, great deals and helpful tips.
The industry has swifty moved beyond building the smartest chatbot to building the most capable AI Agent that becomes the invisible operating layer for your digital life. This includes options such as managing schedules, organizing documents, automating repetitive work, coordinating apps and eventually making decisions with minimal prompting.
A dramatic shift from where AI was even a year ago
The first wave of generative AI was about proving models could generate human-like responses. The next phase is about turning those models into autonomous systems capable of executing tasks across your digital environment.
That’s why terms like “agentic AI,” “subagents” and “ambient intelligence” are suddenly appearing everywhere across Silicon Valley.
With AIG, safety concerns are often top of mind. But Google plans to emphasize safety and says Gemini 3.5 was developed under its Frontier Safety Framework, including expanded cyber and CBRN safeguards designed to reduce harmful outputs while minimizing unnecessary refusals on safe prompts.
That balancing act has become increasingly important as AI systems become more autonomous and capable of taking real-world actions. Additionally, the company also confirmed that Gemini 3.5 Pro — the larger and more capable version of the model family — is already being used internally and is expected to launch publicly next month.
If Gemini 3.5 Flash is about proving speed and responsiveness, Gemini 3.5 Pro may ultimately show how far Google believes autonomous AI systems can actually go.
Bottom line
Gemini 3.5 Flash is generally available today via the Gemini app and AI Mode in Google Search.
With this launch, the company is betting that the future of AI isn’t a tool you constantly open and prompt, but an always-present system quietly managing the chaos of digital life in the background.
And after years of AI being framed as a smarter search box, that may be the biggest shift yet.
Follow Tom's Guide on Google News and add us as a preferred source to get our up-to-date news, analysis, and reviews in your feeds. Subscribe to Tom's Guide on YouTube and follow us on TikTok.
More from Tom’s Guide

Amanda Caswell is one of today’s leading voices in AI and technology. A celebrated contributor to various news outlets, her sharp insights and relatable storytelling have earned her a loyal readership. Amanda’s work has been recognized with prestigious honors, including outstanding contribution to media.
Known for her ability to bring clarity to even the most complex topics, Amanda seamlessly blends innovation and creativity, inspiring readers to embrace the power of AI and emerging technologies. As a certified prompt engineer, she continues to push the boundaries of how humans and AI can work together.
Beyond her journalism career, Amanda is a long-distance runner and mom of three. She lives in New Jersey.
You must confirm your public display name before commenting
Please logout and then login again, you will then be prompted to enter your display name.
