Google Unleashes Gemini 3.5 Flash: AI Model Hits the Streets Fast

This release is part of a broader vision Google calls the "Agentic Gemini Era," aiming for deep AI integration across all its products. Alongside Gemini 3.5 Flash, Google also unveiled Gemini Omni, a new multimodal "world model." Omni focuses on advanced video generation and editing, capable of creating "anything from any input" by combining audio, video, text, and image. Demis Hassabis reportedly compared Omni to progress in artificial general intelligence (AGI). Another significant reveal was Gemini Spark, a new personal AI agent powered by Gemini 3.5, built to proactively take on tasks. During its demonstration, Josh Woodward showed Spark organizing a neighborhood party, handling RSVPs, tracking contributions, drafting emails, and even generating a "hype deck." Pichai lightheartedly told the audience, "Yes, you can close your laptop," while discussing Spark's autonomous capabilities.

Sundar Pichai emphasized Google's strategic focus. He stated, "We've focused on agent decoding, long horizon tasks and real-world workflows." Pichai added, "Gemini 3.5 Flash is very capable model at the frontier but remarkably fast. Flash 3.5 delivers frontier level capabilities at less than half the price and in some cases a third of the price." Koray Kavukcuoglu, CTO of Google DeepMind and Chief AI Architect, further elaborated on its robust capabilities. Kavukcuoglu explained, "[Gemini] 3.5 Flash is especially good when deploying multiple agents simultaneously and completing long-running tasks with massive improvements in coding and tool use. It can independently execute complex coding pipelines or manage iterative research projects entirely by itself. We have even managed to successfully test it by having our agents build a working operating system entirely from scratch."

This wave of innovation extends to Google Search, which is getting a "Generative UI" makeover, with Gemini 3.5 Flash set to become its default model. This new multimodal search box will dynamically expand based on query length and provide AI-generated suggestions. Addressing concerns about AI-generated content, Google is expanding C2PA content credentials and its SynthID watermarking technology to Search and Chrome. Notably, Nvidia, OpenAI, and Elevenlabs will also adopt SynthID, marking a rare instance of collaboration among competitors in the AI space.

For developers and advanced users, Google launched a new $100 AI Ultra plan. This premium offering includes Gemini 3.5 Flash integration for quicker testing, priority access to Google Antigravity, and a substantial 20TB of cloud storage. The company also teased the upcoming release of Gemini 3.5 Pro, an even more advanced version currently in internal testing, expected to launch next month. This latest series of advancements builds on a history of Gemini model development, including versions 1.0, 1.5, 2.0, 2.5, and 3.1 Pro, Flash, and Flash-Lite, each enhancing multimodal understanding, context windows, and agentic actions. The company has been under increasing pressure from rival AI labs such as OpenAI and Anthropic, which have been rapidly deploying new models and features.

Google's I/O 2026 event followed closely on the heels of "The Android Show" last week, where the company previewed Android 17 updates heavily integrating AI, further cementing Google's strategy to embed AI throughout its entire product ecosystem. This aggressive push with Gemini 3.5 Flash and its related models signals a significant shift in how users and developers will interact with AI, emphasizing speed, autonomy, and broad integration into daily digital life.