Google Releases Gemini 3.5 Flash at I/O 2026: Everything You Need to Know

Liputan6.com, Jakarta - Google officially released Gemini 3.5 Flash on May 19, 2026, at its annual Google I/O 2026 developer conference.

This model is the first in the Gemini 3.5 series, specifically designed to deliver advanced intelligence at higher speeds and lower costs.

Its primary focus is on supporting complex agent and coding tasks, marking a major step forward in the evolution of AI.

Gemini 3.5 Flash promises not only superior performance but also significant efficiency gains, making it an attractive choice for a wide range of applications.

With advanced multimodal capabilities and a large context window, Gemini 3.5 Flash is ready to meet the challenges of modern computing.

Starting today, Gemini 3.5 Flash is rolling out to everyone globally, free of charge.Try it by selecting “3.5 Flash” from the model dropdown menu at https://t.co/382WL5xSvc or in the app.

— Google Gemini (@GeminiApp) May 20, 2026

Gemini 3.5 Flash is now globally available starting May 19, 2026, making it accessible to billions of people worldwide.

For general users, it comes standard in the Gemini app and AI Mode in Google Search.

Developers can access it through Google AI Studio, Android Studio, Gemini API, Google Antigravity, and the Gemini Enterprise Agent Platform.

For enterprises, Gemini 3.5 Flash is available on the Gemini Enterprise Agent Platform and Gemini Enterprise, expanding its reach across various sectors.

This model is claimed to be four times faster in outputting tokens per second (TPS) than other frontier models.

Gemini 3.5 Flash also outperforms Gemini 3.1 Pro on most coding and agent benchmarks, including Terminal-Bench 2.1 (76.2%), MCP Atlas (83.6%), and CharXiv Reasoning (84.2%).

Gemini 3.5 Flash supports text, image, audio, and video input, with text output, demonstrating robust multimodal capabilities.

Furthermore, Gemini 3.5 Flash supports an input context window of 1 million tokens and a maximum output window of 65,536 tokens, enabling the processing of very large amounts of information.

In terms of cost, Gemini 3.5 Flash is priced at $1.50 per 1 million input tokens and $9.00 per 1 million output tokens at the standard level.

Google states that using this model can provide significant cost savings compared to other frontier models, although it is more expensive than previous Flash models.

The knowledge limit for this model is January 2026.

However, Gemini 3.5 Flash does not support audio creation, computer usage, image creation, or the live API.

This model has the "Thinking" feature enabled by default, with a "medium" effort level for good, faster, and more cost-effective results.

Gemini 3.5 Flash is no longer positioned solely as a lightweight, low-cost AI platform; it is now approaching the performance of the Pro line, previously considered the highest-end.

While more expensive than previous Flash models, Gemini 3.5 Flash offers frontier-level intelligence at up to four times the speed and often at less than half the cost.

Google sees the future of AI not as a tool that is constantly being opened and prompted, but as a system that is always present, quietly managing the chaos of digital life in the background.

The company is focusing on "agentic workflows," where AI can understand tasks, plan steps, and complete tasks like a digital assistant.

In addition to Gemini 3.5 Flash, Google also announced that a new frontier AI model, Gemini 3.5 Pro, will be released in June.

The Gemini 3.5 Pro model is currently being used internally by Google for various development projects.

The arrival of this Pro version is expected to bring even more advanced capabilities in the future.

Another exciting innovation is Gemini Omni, a new world model that represents a major leap forward in world understanding, multimodality, and editing.

Google also introduced Gemini Spark, a 24/7 personal AI agent designed to proactively manage tasks and help users navigate their digital lives.

Gemini Spark will be rolled out first to trusted testers, with a beta planned for Google AI Ultra subscribers in the US soon.