Google launches Gemini 3.5 Flash, Gemini Omni, with faster AI performance at I/O 2026

Google has officially unveiled Gemini 3.5 Flash during the Google I/O 2026 keynote, introducing what the company describes as a faster and smarter AI model designed for coding, reasoning and advanced agentic tasks.

According to Google, Gemini 3.5 Flash is built to deliver frontier-level AI performance while maintaining Flash-level speed and latency. CEO Sundar Pichai said the model performs four times faster than several other frontier AI models in terms of output token generation.

Google says speed no longer comes at the cost of intelligence

Google heavily emphasized that Gemini 3.5 Flash is designed to combine speed and intelligence together instead of sacrificing one for the other.

The company claims the model rivals large flagship AI systems across multiple benchmarks while still delivering significantly faster responses. Google also stated that Gemini 3.5 Flash outperforms Gemini 3.1 Pro across several coding and agentic AI benchmarks.

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

It delivers frontier-level performance at 4x the speed of comparable frontier models — often at less than half the cost.

Generally available, starting today. 🧵#GoogleIO pic.twitter.com/jLhqozutwG

— Google (@Google) May 19, 2026

Gemini 3.5 Flash focuses on coding and reasoning

Google positioned Gemini 3.5 Flash as an AI model designed for complex real-world tasks including advanced reasoning, iterative coding and long-horizon agentic workflows.

The company said the model is best suited for:

• Everyday AI tasks
• Agentic coding
• Advanced reasoning
• Multimodal understanding
• Long-context processing

Google also explained that Gemini 3.5 Flash can support teams of AI subagents and generate richer interactive web interfaces and graphics.

Gemini 3.5 Flash supports multimodal AI experiences

Another major highlight is multimodal understanding. Google said the model can process and understand text, audio, images, video and code simultaneously.

The company described Gemini 3.5 Flash as a truly multimodal AI system capable of handling complex workflows across different formats and media types.

Gemini 3.5 Flash rolling out across Google ecosystem

Google confirmed that Gemini 3.5 Flash is available starting today across multiple platforms and services.

The rollout includes:

• Gemini app
• Gemini API
• Google AI Mode
• Google AI Studio
• Android Studio
• Gemini Enterprise platforms

Google also confirmed that a more advanced flagship model called Gemini 3.5 Pro is currently in development and expected to launch soon

Gemini Omni

Google also introduced Gemini Omni during the Google I/O 2026 keynote, a new multimodal AI video model capable of generating editable videos using text, images, audio and existing video clips. Unlike earlier AI video tools focused mainly on text-to-video generation, Gemini Omni allows users to modify scenes, characters and visual elements through conversational text prompts.

Google said the model comes with a stronger understanding of real-world physics including motion, gravity and fluid behaviour to make AI-generated videos appear more realistic. The company also confirmed that Gemini Omni outputs will include SynthID digital watermarks to address misuse concerns.

Add Techlusive as a Preferred Source

The first version called Gemini Omni Flash is rolling out through the Gemini app, Google Flow and YouTube Shorts, with free access available for Shorts users while advanced access remains limited to Google AI Plus, Pro and Ultra subscribers.