We designed Gemini 2.5 as a household of hybrid inference fashions that ship unimaginable efficiency. Pareto frontier relationship between price and velocity. At this time, we’re taking the following step by making the two.5 Professional and Flash fashions typically accessible as secure variations. And we’re now providing a preview of two.5 Flash-Lite, our most cost-effective and quickest 2.5 mannequin up to now.
2.5 Flash and a pair of.5 Professional now typically accessible
Because of your suggestions, at this time we’re releasing secure variations of two.5 Flash and Professional. This lets you construct manufacturing purposes with confidence. developer splines, rooms, and so on. organization Firms like Snap and SmartBear have already been utilizing the newest variations in manufacturing for the previous few weeks.
Introducing Gemini 2.5 Flash-Lite
We additionally preview the brand new Gemini 2.5 Flash-Lite, our most cost-effective and quickest 2.5 mannequin up to now. You can begin constructing with the preview model at this time. We welcome your suggestions.
2.5 Flash-Lite has larger total high quality than 2.0 Flash-Lite in coding, math, science, reasoning, and multimodal benchmarks. It excels at high-volume, delay-sensitive duties reminiscent of translation and categorization, and has decrease latency than 2.0 Flash-Lite and a pair of.0 Flash over a variety of immediate samples. Gemini 2.5 has the identical options that make Gemini 2.5 helpful, together with the flexibility to allow considering on completely different budgets, connectivity to instruments like Google search and code execution, multimodal enter, and a context size of 1 million tokens.
Study extra in regards to the newest 2.5 household fashions. Gemini Technical Report.
