Google has expanded its Gemini family of artificial intelligence models with the launch of three new versions aimed at developers, businesses and cybersecurity professionals. The company introduced Gemini 3.6 Flash, Gemini 3.5 Flash Lite and Gemini 3.5 Flash Cyber, while also announcing that the release of the much-awaited Gemini 3.5 Pro has been delayed for additional improvements.
The announcement highlights Google’s strategy of building AI models tailored for different user needs instead of offering a single solution for every task. The latest models focus on faster performance, lower operating costs and specialised capabilities that can support businesses deploying AI across multiple sectors.
Leading the new lineup is Gemini 3.6 Flash, Google’s latest high-performance AI model. According to the company, it offers better reasoning, coding, instruction-following and multimodal capabilities while maintaining fast response times. The model is designed to automatically adjust its reasoning depending on the complexity of a prompt, allowing it to respond quickly to simple requests while using additional computing power for more challenging tasks.
Google says this adaptive approach helps developers achieve higher-quality results without significantly increasing costs or slowing down applications. The model is expected to be useful for AI assistants, software development, content creation and enterprise automation.
The company has also launched Gemini 3.5 Flash Lite, which it describes as its most affordable AI model so far. Built for organisations handling large volumes of AI requests, Flash Lite is designed to deliver reliable performance at lower cost.
The model is suitable for everyday business applications such as document summarisation, customer support, language translation, data classification and content generation. By reducing operating expenses, Google hopes the model will appeal to startups and enterprises looking to expand AI adoption without increasing infrastructure costs.
A major addition to the Gemini family is Gemini 3.5 Flash Cyber, a model developed specifically for cybersecurity. Unlike general-purpose AI systems, Flash Cyber has been trained using threat intelligence, vulnerability databases and security research.
The model can assist security teams in analysing malware, identifying vulnerabilities, investigating cyber threats and supporting incident response. Google believes specialised AI will become increasingly important as cyberattacks grow more sophisticated and organisations face greater pressure to strengthen digital security.
The launch reflects growing demand for AI systems that can perform specialised tasks instead of relying only on broad language capabilities. Enterprises are increasingly seeking AI models designed for coding, security, customer service and workflow automation.
While introducing the new Flash models, Google also confirmed that Gemini 3.5 Pro will not be released as originally planned. The company said it requires additional testing to improve reliability and overall performance before making the model publicly available.
Although Google did not announce a revised release date, it said the delay would help ensure a better experience for developers and enterprise customers. The decision comes as AI companies face increasing expectations around model quality, safety and accuracy.
The newly launched models are available through Google AI Studio, the Gemini API and Vertex AI, allowing developers and businesses to integrate them into their applications immediately. This gives organisations access to AI tools for coding, automation, customer engagement, cybersecurity and business operations.
The expansion also reflects the intense competition in the global AI industry, where companies are racing to introduce faster and more capable models. Instead of focusing only on larger AI systems, firms are increasingly developing specialised models that offer better performance for specific business needs.
Google said the new Gemini Flash models are designed to provide a balance between speed, affordability and accuracy, enabling developers to build AI-powered products more efficiently. The company expects the latest releases to support wider AI adoption across industries ranging from technology and finance to healthcare, education and cybersecurity.
With three new Gemini models now available and the Pro version still under development, Google is continuing to broaden its AI ecosystem while focusing on practical applications for developers and enterprise customers.
Also Read: Agilus appoints Vijender Singh as MD & CEO