Google’s new Gemini AI model and tools are all about agents now


Google I got new Gemini model of artificial intelligenceIt is designed to be a proxy. The company debuted Gemini 3.5 Flash on Tuesday in Google I/O Annual developer conference, with the more powerful Gemini 3.5 Pro expected next month.

Google says its new AI models are comparable to others Border and near-border models – Like OpenAI GPT-5.5 Or anthropic Closing of business 4.7 – But also focus on being faster and more capable of handling tasks for you.

You’ll be able to use these templates in many different ways, including It’s similar to OpenClaw 24/7 agent assistant and upgraded anti-gravity coding platform. Here’s what Gemini’s coming from Google I/O.

New pair of Gemini models

The first new model, Gemini 3.5 Flash, is rolling out to all users starting now. Google CEO Sundar Pichai told reporters during a pre-I/O press conference that the model is “incredibly fun to use” and a “game-changer” internally at Google. He said it has been trained to work quickly and complete tasks at about half the cost of competing models.

Gemini 3.5 Flash is particularly good at running agents — autonomous AI instances that can handle certain tasks. This involves running multiple agents at the same time to handle larger, long-term projects. He can handle multi-hour sessions on his own, and complete entire coding research projects, said Koray Cavukoglu, chief technology officer at Google DeepMind and chief AI engineer at Google.

Still to come is the Gemini 3.5 Pro, which Pichai said Google is testing internally.

Spark: a 24/7 AI assistant

In what appears to be a competitor to OpenClaw, the autonomous AI agent platform It exploded at the beginning of the yearGoogle is preparing an assistant called Gemini Spark. Josh Woodward, vice president of Google Labs, Gemini App and AI Studio, said the company will launch Spark “deliberately,” with a group of trusted users getting access this week, followed by a beta for AI Ultra subscribers next week.

Spark is an always-on AI agent that can act on your behalf and follow directions. It’s in the Google cloud, so there’s no need to buy one Mac Minis are disappearing fast Runs OpenClaw – and can perform tasks in Workspace apps like Gmail and Docs.

Screenshots of Gemini Spark game with to-do list.

The Google Gemini Spark 24/7 AI agent can access things like your Gmail files and Docs to take actions on your behalf.

Google

“When you use it, it’s like you’re throwing things over your shoulder, and Spark picks them up and gets the job done,” Woodward said.

Google testers use it to plan parties, track school schedules and monitor inboxes for questions, Woodward said. Google is planning more connections over the summer so Spark can run a wide range of third-party apps and websites through Chrome.

Anti-gravity expansion

Coding is one area where Google has fallen behind rivals Anthropic and OpenAI, which have thrived with CloudCode and… Codex Alimentariusrespectively. Now Google has reimagined its platform, Antigravity, as an agent-first development tool. However, Google’s vision for antigravity goes beyond just programming. The new direction is “a platform for developing and managing teams of autonomous AI agents,” Cavukoglu said.

This includes a standalone desktop application for Antigravity to serve as a home for the agent AI. The tasks of these agents can vary. Kavukcuoglu gave the example of one agent creating a website, another creating brand assets, and a third planning products. The Antigravity feature will also be available on the command line.

Gemini app updates

With all these powerful changes in Google’s AI products, expect a major transformation for Gemini. Woodward said the application will include the new Gemini 3.5 Flash and… Gemini Omni models, along with some new features.

The new design, called Neural Expressive, includes a new set of animations, colors, and haptic feedback when you tap buttons. Woodward said Gemini will also respond with more images, timelines or visualizations when appropriate.

An image of the Gemini app on a phone, with the AI ​​being asked to tell the user "Ask away, Camille!"

Google’s Expressive Neural Design Language is the new interface for Gemini.

Google

Another feature called Daily Brief is a personalized presentation of information that Gemini’s AI will collect based on your inbox, calendar tasks, and other data. It will be available this week for Google AI Plus, Pro, and Ultra subscribers In the United States. Woodward called it “a great way to introduce people to the power of customers.”



Leave a Reply

Your email address will not be published. Required fields are marked *