Google Launches Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber
Google released three new Gemini models with better efficiency, lower costs, and a cybersecurity-focused variant for vulnerability detection.
BY FOUNDERBUILT AI NEWS
Google has released three new Gemini models: Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber. The flagship 3.6 Flash delivers a step up in coding and knowledge work while consuming 17 percent fewer output tokens than 3.5 Flash, making it more cost-effective for production AI agents at USD 1.50 per 1M input and USD 7.50 per 1M output tokens.
The new Gemini 3.5 Flash-Lite is Google's fastest model yet at 350 tokens per second. Priced at USD 0.30 per 1M input and USD 2.50 per 1M output tokens, it outperforms the previous 3 Flash on agentic benchmarks like SWE-Bench Pro and OSWorld-Verified. It also supports computer use as a built-in tool for agentic tasks.
Google also introduced Gemini 3.5 Flash Cyber, a specialized model fine-tuned for cybersecurity vulnerability detection and patching. It will initially be available exclusively to governments and trusted partners through the CodeMender platform. Meanwhile, Gemini 3.5 Pro is testing with partners, and Google has begun its most ambitious pre-training run yet for Gemini 4.