Google has introduced a new generation of Gemini artificial intelligence models aimed at helping individuals, developers and businesses complete complex digital tasks more efficiently.
The announcement, made on 21 July 2026, includes Gemini 3.6 Flash, Gemini 3.5 Flash-Lite and Gemini 3.5 Flash Cyber. Google says the models are designed to provide improved speed, reliability and cost efficiency, particularly for organisations building AI-powered agents and automated workflows.
Gemini 3.6 Flash is positioned as the main model for demanding everyday tasks. It offers improvements in coding, knowledge-based work, document processing, chart analysis and other activities that require the AI to understand different types of information. Google says the model uses fewer output tokens than Gemini 3.5 Flash, which could help developers reduce the cost of operating AI applications.
The company also launched Gemini 3.5 Flash-Lite, a faster and lower-cost model intended for high-volume tasks such as search, document processing, translation and automated customer services. It is designed for organisations that need to process many requests quickly while controlling operational costs.
Gemini 3.5 Flash Cyber is focused on cybersecurity. Google says the specialised model can help identify, analyse and fix software vulnerabilities. Because cybersecurity tools can also be misused, the model will initially be available only to governments and selected trusted partners through a limited-access programme.
Gemini 3.6 Flash and Gemini 3.5 Flash-Lite are available through Google AI Studio, the Gemini API, Android Studio, Gemini Enterprise and the Gemini application. Flash-Lite is also being introduced gradually in Google Search.
The announcement demonstrates the growing competition among major technology companies to provide faster and more affordable artificial intelligence systems for consumers, developers and businesses around the world.
