Google launched Gemini 2.5 Flash, cheap AI model will be used in these works

Google has introduced Gemini 2.5 Flash, the second in its Gemini 2.5 model family. It is an affordable, low-latency (quick response rate), and suitable model for real-time inference, big conversation, and general-purpose use. The new AI model will be made available on both Google AI Studio and Vertex AI platforms in the near future to enable developers and users to leverage it to develop applications and AI agents.

Google released details regarding its newest Large Language Model (LLM), on its blog. In the blog, the company mentioned that the Gemini 2.5 Pro model has also become available on Vertex AI. The Pro model is ideal for deep knowledge-intensive tasks with multi-step logic and sophisticated decision-making. The Flash model is made with speed, latency reduction, and affordability in mind instead.

Google describes the Flash model as a “workhorse”. It is an ideal engine for virtual assistants, real-time summary tools, etc. where fast and accurate results are required at a large scale.