Google releases three new Gemini models — but no 3.5 Pro

Editor's take

Google has launched three new, faster Gemini models: Gemini 3.6 Flash, 3.5 Flash-Lite, and Flash Cyber, designed for efficiency and speed. The omission of the previously announced Gemini 3.5 Pro from this release, following its delayed availability, suggests a strategic pivot or internal recalibration within Google's AI development timeline.

This development is significant because Gemini 3.5 Pro was positioned as a major advancement, particularly its claimed one-million-token context window, which would have offered a substantial advantage over competitors like OpenAI's GPT-4 Turbo. The delay and focus on these new "Flash" variants indicate a potential prioritization of cost-effectiveness and broad deployment over raw capability for now, impacting developers and businesses relying on rapid access to cutting-edge LLM features.

Future attention should focus on the actual performance benchmarks of these new Flash models compared to existing offerings and the eventual rollout timeline for Gemini 3.5 Pro. Understanding the specific use cases these Flash models are optimized for, and whether the one-million-token capability of 3.5 Pro will be integrated into a more accessible model, will clarify Google's competitive positioning in the LLM market.