Google on Monday, July 21, introduced three new Gemini artificial intelligence models aimed at improving the speed, efficiency and affordability of AI applications, as the company seeks to strengthen its position in the increasingly competitive generative AI market.
The new releases include Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, and Gemini 3.5 Flash Cyber, each designed for different production workloads ranging from coding assistants to high-volume automation and cybersecurity.
Google said Gemini 3.6 Flash improves on its predecessor by reducing output token usage by 17% while delivering stronger coding, multimodal reasoning and knowledge work performance at a lower cost. The company said some coding benchmarks recorded token reductions of up to 65%, allowing developers to build AI agents more efficiently.
Alongside it, Google released Gemini 3.5 Flash-Lite, describing it as its fastest and most cost-effective model in the 3.5 family. The model is designed for high-throughput applications such as document processing, search and large-scale automation, producing up to 350 output tokens per second while offering configurable reasoning levels for different workloads.
Google also announced Gemini 3.5 Flash Cyber, a specialized version fine-tuned for cybersecurity. Integrated with Google's CodeMender security agent, the model is designed to identify, validate and patch software vulnerabilities. Due to the technology's dual-use nature, the company said the model will initially be available only to governments and selected trusted partners through a limited-access pilot.
The company said Gemini 3.6 Flash and Gemini 3.5 Flash-Lite are available immediately through Google AI Studio, the Gemini API, Android Studio, enterprise platforms and the Gemini app, with Flash-Lite also rolling out in Google Search.
Google added that Gemini 3.5 Pro remains in partner testing and will be released broadly once it is ready. The company also confirmed that development is underway on Gemini 4, which it described as its most ambitious pre-training effort to date.
I truly appreciate you spending your valuable time here. To help make this blog the best it can be, I would love your feedback on this post. Let me know in the comments: How could this article be better? Was it clear? Did it have the right amount of detail? Did you notice any errors?
If you found any of the articles helpful, please consider sharing it.