Gemini 3.5 Flash is the first model in the new Gemini 3.5 family — a line designed to combine high capability with real-world action execution. It is built for complex multi-step tasks and extended workflows, and Google claims it generates four times more tokens per second than other advanced models. A Gemini 3.5 Pro version is promised for next month.

Meet Gemini 3.5 Flash — our strongest agentic and coding model yet.
Meet Gemini 3.5 Flash — our strongest agentic and coding model yet. X

Google is framing the launch as a shift from conversational AI to agentic AI — systems that don't just answer questions but execute tasks sequentially. For certain actions such as sending emails or making financial transactions, the agent will request additional user permission before proceeding.

Gemini 3.5 Flash performance in major benchmarks. Source: Google.
Gemini 3.5 Flash performance in major benchmarks. Source: Google.

Gemini 3.5 Flash is being integrated into the Gemini app and other Google products. The company's developer blog states the model is designed to "accelerate the transition from prompts to actions."

Google also introduced Gemini Spark — a personal AI assistant powered by Gemini 3.5 that handles everyday tasks and integrates freely with Google's services.

Antigravity 2.0

Antigravity 2.0 serves as a central platform for agent coordination, enabling multiple digital assistants to work in parallel across different workflows. It supports dynamic sub-agents, scheduled background automation tasks, and deep integration with Google AI Studio, Android, and Firebase.

New additions include Antigravity CLI for instant agent creation without a graphical interface, Antigravity SDK for programmatic access to the same agent toolset used in Google's own products, and Antigravity on the Gemini Enterprise Agent Platform — allowing Google Cloud customers to connect the service directly to their projects.

A new Google AI Ultra plan has been introduced at $100 per month, offering five times the Antigravity usage limit of Google AI Pro.

Gemini Omni

Google also unveiled Gemini Omni — a model designed to turn creative visions into reality. It natively combines text, images, and video to generate high-quality cinematic content with minimal effort. Users can apply zoom effects and swap backgrounds using simple text prompts.

Google Pics

Google Pics is a new image creation and editing tool capable of producing everything from party posters to infographics. It automatically identifies objects in photos and analyzes their relationships for more intelligent editing.

Managed Agents in the Gemini API

With a single API call, developers can now launch a managed agent capable of logical reasoning, tool use, and code execution in an isolated Linux environment. These agents run on the Antigravity platform, built on Gemini 3.5 Flash, and are accessible via the Interactions API and Google AI Studio.

We’re partnering with @Samsung, @_GentleMonster_ and @WarbyParker on new intelligent eyewear.
We’re partnering with @Samsung, @_GentleMonster_ and @WarbyParker on new intelligent eyewear. X

Google also confirmed it is working on smart glasses and previewed the design of an upcoming autumn collection.

Conclusion:

Gemini 3.5 Flash signals a decisive shift from conversational AI to autonomous execution — and with Antigravity providing the coordination infrastructure, Google is building an entire operating layer for the agentic era, not just a faster model. If Gemini 3.5 Pro delivers on its promise next month, Google will be in a strong position to challenge OpenAI and Anthropic in the enterprise segment.