Google has introduced a new addition to its Gemini AI Model Line-Up-Zemini 2.5 flash-light. According to Google, this new AI model can give the lowest cost and fastest high performance. With the new model, the company has announced the general availability of Gemini 2.5 flash and Pro model for all users.

Google says that Gemini 2.5 Flash-Light 2.5 is its cheapest and fastest model in the family. It is designed to handle large versions of delay-sensitive functions such as translation, classification and argument on low computational costs. Compared to its predecessor, 2.0 flash-light, new models are asked to provide better accuracy and quality in coding, mathematics, science, logic and multimodal benchmarks. “It excels in high-volume, delay-sensitive functions such as translations and classifications, with a wide sample of signals with low delay and 2.0 flash than 2.0 flash-light,” says Google.

In spite of being light that google has been highlighted, 2.5 flash-light comes with a full suit of advanced capabilities. These include support for multimodal inputs, integration with devices such as 1 million-token reference window, Google search and code execution, and flexibility to modify computational thinking based on budget. According to the company, these features make Gemini 2.5 flash-light ideal for developers who are looking to balance efficiency with strong AI functionality.

Gemini 2.5 flash-light availability

The Gemini 2.5 flash-light model is currently available in preview through Google AI Studio and Wartax AI. Google has also integrated optimized versions of 2.5 flash-lights and expanding their access to everyday users beyond the developers, such as search-like discovery in their main products.

Gemini 2.5 Flash and Pro models are now available to all

In addition to starting a flash-light, Google has also announced that its Gemini 2.5 flash and Gemini 2.5 Pro models are now stable and generally available. These models were earlier accessible to a select group of developers and organizations for early production use.

According to Google, companies such as SNAP, Smartbear, and Creative Tools Provider Spine have already integrated into their workflows with the results that encourage these models. Now when flash and pro are completely open, developers can use them in production-grade applications with more confidence.

Both stable and preview models can be accessed through Google AI Studio, Wartax AI and Gemini apps.