On Tuesday, Google unveiled Gemini 3.5 Flash, a new AI model it claims is its most powerful yet for coding and autonomous AI agents. Introduced at the company’s annual I/O developer conference, this model can independently execute coding pipelines, manage research projects, and, according to internal tests, build an entire operating system from scratch. This release marks Google’s strategic pivot from positioning AI as a conversational tool to a more agentic one—not just answering questions, but planning, building, and iterating on real tasks with minimal human involvement.

“3.5 Flash offers an incredible combination of quality and low latency,” Koray Kavukcuoglu, DeepMind’s chief technologist, told reporters on Monday ahead of the public launch. “It outperforms our latest frontier model, 3.1 Pro, on nearly all the benchmarks,” including coding, agentic tasks, and multimodal reasoning. He added that it is four times faster than other frontier models, a speed ideal for coding and agentic tasks. However, Google has “taken it to another level” by developing an optimized version of Flash that is 12 times faster while maintaining the same quality. According to Kavukcuoglu, this speed is central to Flash’s design for agentic work, where multiple AI agents run simultaneously on long-running tasks. On stage at I/O, Google engineer Varun Mohan demonstrated agents spawning off to work on separate components before coming together to build a full operating system inside Antigravity, the company’s agentic development platform and IDE.

Kavukcuoglu noted that Flash 3.5 was co-developed with Antigravity to provide agents with a “native environment where they can live, work, and execute.” At I/O, Google released Antigravity 2.0, a standalone desktop application designed around agent-first development.

The gains extend beyond demos. Google reports that 3.5 Flash’s agentic capabilities are already making an impact among partners, such as banks and fintechs automating multi-week workflows, and data science teams uncovering insights in complex data environments. The model can run autonomously for several hours, though Tulsee Doshi, Google’s senior director and head of product, noted that it will occasionally pause and request user input when it encounters a decision point or permission issue requiring human judgment. When Google releases its forthcoming 3.5 Pro model, the two are designed to work in tandem. “I think it really comes down to where you actually want that reasoning power—where you want that larger model that can really push on the reasoning side—versus where you have tasks that really merit good brute force tool use capabilities,” Doshi explained.

3.5 Flash is now the default model in the Gemini app and in AI Mode in Search globally. At I/O, Google also announced agentic capabilities coming to Search, allowing users to create, customize, and manage AI agents directly on the platform. The new model will also power Gemini Spark, Google’s new personal AI agent designed to run 24/7 to help consumers manage their digital lives.

Providing such advanced AI capability to average consumers invites scrutiny. Google is currently facing a lawsuit after a man nearly committed a mass casualty event and died by suicide following weeks of chatting with Gemini last year. The potential for harm only grows when making powerful autonomous agents more broadly available. Google states that Gemini 3.5 has strengthened cyber and CBRN (Chemical, Biological, Radiological, and Nuclear) safeguards and is better calibrated to engage with sensitive questions rather than refuse them outright. Gemini 3.5 Flash is available generally today via Antigravity, the Gemini API, and Gemini Enterprise, as well as through the Gemini app and AI mode in Search.