Google DeepMind Unveils Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber

Google DeepMind has introduced three new models in the Gemini family. The 3.6 Flash model, built on feedback from 3.5 Flash, delivers better coding and knowledge work quality while consuming 17% fewer output tokens (per the Artificial Analysis index) and at a lower price: $1.50 per million input tokens and $7.50 per million output tokens. In benchmarks, the model shows improvements: DeepSWE — 49% versus 37% for 3.5 Flash, MLE Bench — 63.9% versus 49.7%, OSWorld-Verified — 83.0% versus 78.4%. The model incorporates enhanced safety measures in the areas of CBRN (chemical, biological, radiological, nuclear weapons) and cyber attacks. The 3.5 Flash-Lite model is the fastest in the 3.5 series — 350 output tokens per second (per Artificial Analysis) — and the most cost-effective: $0.3 per million input tokens and $2.5 per million output tokens. It significantly outperforms 3.1 Flash-Lite in quality, for example, in Terminal-Bench 2.1 (54% versus 31%), GDM-MRCR v2 (72.2% versus 60.1%), and GDPval-AA v2 (1140 versus 642). The 3.5 Flash Cyber model, based on 3.5 Flash and fine-tuned for cybersecurity, works as part of the CodeMender agent and achieves competitive results on the CyberGym benchmark. It will be available only to governments and trusted partners through a limited pilot access. The 3.6 Flash and 3.5 Flash-Lite models are now available to developers in the Gemini API via Google AI Studio and Android Studio, as well as on the enterprise platform Gemini Enterprise and the Gemini app.