Google Unveils New Gemini AI Models to Cut Enterprise Costs and Speed Up AI Development

Google DeepMind has introduced new Gemini AI models focused on faster performance, lower operating costs and improved enterprise productivity, while confirming its next flagship model is still in testing.
Uche Emeka
Uche EmekaAI9 hours ago2 minute read
Key Points
Google DeepMind has launched new Gemini models, 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber, focusing on efficiency and cost reduction for autonomous software agents.
Gemini 3.6 Flash is optimized for coding and knowledge work with improved performance and token efficiency, while 3.5 Flash-Lite offers high speed and cost-effectiveness for high-volume tasks.
The specialized Gemini 3.5 Flash Cyber is designed for validating and remediating security flaws, with its distribution restricted to governments and vetted partners.
Google Unveils New Gemini AI Models to Cut Enterprise Costs and Speed Up AI Development

Google DeepMind has expanded its artificial intelligence portfolio with the launch of Gemini 3.6 Flash, Gemini 3.5 Flash-Lite and Gemini 3.5 Flash Cyber, a new lineup designed to help businesses build faster, more affordable AI-powered applications.

The company said the latest models focus on improving efficiency, reducing latency and lowering computing costs for organizations deploying AI agents at scale.

Faster AI for Businesses

Google positioned Gemini 3.6 Flash as its primary model for coding, document analysis and multimodal reasoning, claiming it delivers stronger performance while using fewer output tokens than its predecessor.

The company says the model has already been adopted by organizations including Figma, Harvey and Hebbia to accelerate software development, legal research and document processing.

The goal is to reduce the cost of running AI systems without sacrificing performance, making advanced AI more practical for enterprise workloads.

Flash-Lite Targets High-Volume Tasks

Google also introduced Gemini 3.5 Flash-Lite, a lower-cost model designed for high-volume tasks such as document processing, search and automated workflows.

According to Google DeepMind, the model is optimized for speed and affordability, enabling developers to reserve more powerful AI models for complex tasks while routing routine requests to Flash-Lite.

The company says this approach can significantly reduce infrastructure costs for organizations running AI applications at scale.

New Cybersecurity Model Released

Alongside the productivity-focused models, Google unveiled Gemini 3.5 Flash Cyber, a specialized AI system built to help identify and remediate software security vulnerabilities.

Access to the cybersecurity model will initially be limited to governments and selected partners through a pilot programme, with Google citing safeguards designed to reduce the risk of misuse.

Gemini Pro Still on the Horizon

Notably absent from the announcement was a new version of Google's flagship Gemini Pro model.

Google confirmed that the next-generation Gemini Pro remains in partner testing and is expected to launch soon, while work has already begun on the company's future Gemini 4 architecture.

The latest releases underscore Google's strategy of making AI systems faster, more efficient and more affordable as competition intensifies across the global artificial intelligence market.

Loading...