Google has revealed the general release of its Gemini 2.5 Flash-Lite model, signifying the conclusion of its 2.5 series. This model is acclaimed as Google’s quickest and most economical AI model within the series, with pricing established at $0.10 per 1M token input and $0.40 for output. The Gemini 2.5 Flash-Lite, which has transitioned out of its preview phase, has been trialed in practical situations, showcasing its skills in programming, mathematics, science, reasoning, and multimodal comprehension.
Organizations such as Satlyt have adopted the model to improve telemetry data summarization and satellite data management, realizing a 30% decrease in energy usage. HeyGen has harnessed the model to translate video content into more than 180 languages. Other enterprises, including DocsHound and Evertune, have employed it to speed up video processing and report creation.
The Gemini 2.5 Flash-Lite model was first introduced in a preview phase in June, together with the public launch of the 2.5 Flash and 2.5 Pro models. Developers can now utilize the Gemini 2.5 Flash-Lite in Google’s AI Studio and Vertex AI by inputting “gemini-2.5-flash-lite” in their code. This launch follows a sequence of Gemini 2.5 models, with Google highlighting the model’s rapidity and affordability for developers aiming to enhance their workflows.
