With Gemini 3.5 Flash, Google bets its next AI wave on agents, not chatbots

Source: TechCrunch

Launch and capabilities

Google launched Gemini 3.5 Flash, a new AI model described as its strongest yet for coding and autonomous AI agents. Introduced at the company’s annual I/O developer conference, the model can independently execute coding pipelines, manage research projects, and, in internal tests, build an operating system from scratch.

The release signals Google’s shift from positioning AI as a conversational tool to an agentic one—planning, building, and iterating on real work with minimal human input.

“3.5 Flash offers an incredible combination of quality and low latency,” said Koray Kavukcuoglu, DeepMind’s chief technologist. “It outperforms our latest frontier model, 3.1 Pro, on nearly all benchmarks, including coding, agentic tasks, and multimodal reasoning.”

Performance and speed

Kavukcuoglu noted that Flash is four times faster than other frontier models, a speed ideal for coding and agentic tasks. Google has also developed an optimized version that is 12 times faster while maintaining the same quality.

The speed is central to Flash’s design for agentic work, where multiple AI agents run concurrently on long‑running tasks. At I/O, Google engineer Varun Mohan demonstrated agents spawning off to work on separate components before assembling a full operating system inside Antigravity, Google’s agentic development platform and IDE.

Image credit: Google

Kavukcuoglu added that Flash 3.5 was co‑developed with Antigravity so agents have a “native environment where they can live, work, and execute.”

Agentic development platform: Antigravity

At I/O, Google released Antigravity 2.0, a standalone desktop application built around agent‑first development. The platform enables agents to operate autonomously for multiple hours, though Tulsee Doshi, Google’s senior director and head of product, explained that agents may pause and request user input when they encounter a decision point or permission issue requiring human judgment.

When Google later releases the forthcoming 3.5 Pro model, the two are designed to work in tandem:

• 3.5 Pro acts as the orchestrator and planner.
• Flash serves as the sub‑agents that execute tasks.

“3.5 Pro becomes your orchestrator, your planner, and then it actually can leverage Flash to be the various sub‑agents,” Doshi told TechCrunch.

Impact and use cases

Google reports that Flash’s agentic capabilities are already creating impact among partners, such as banks and fintechs automating multi‑week workflows, and data‑science teams extracting insights from complex data environments.

The model is now the default in the Gemini app and in AI Mode in Search worldwide. Google also announced that agentic capabilities will be coming to Search, allowing users to create, customize, and manage AI agents directly on the platform. Additionally, Flash will power Gemini Spark, a 24/7 personal AI assistant integrated with Gmail to help consumers manage their digital lives.

Safety and concerns

Providing such powerful AI capabilities to average consumers raises scrutiny. Google is currently facing a lawsuit after a man, who had been chatting with Gemini, died by suicide following weeks of interaction. The company acknowledges that the implications for harm grow when powerful autonomous agents become broadly available.

Google says 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.

Availability

Gemini 3.5 Flash is available today via:

• Antigravity
• Gemini API
• Gemini Enterprise
• Gemini app
• AI Mode in Search