Google has announced the public preview of Gemini 1.5 Pro on its Vertex AI platform, marking a significant advancement in generative AI capabilities. The model, which supports up to 1M tokens, is designed to power new features in Code Assist and other applications. Gemini 1.5 Pro, described as Google's most capable generative AI model to date, was launched in February and is capable of processing between 128,000 and 1M tokens. This release is part of Google's broader effort to integrate AI more deeply into its cloud services, as highlighted at the Google Cloud Next event. The model now includes the ability to process audio streams, including speech, and has been enhanced with a new File API for easier file handling. Additionally, Google has introduced the ability to enhance Gemini models with Google Search, allowing businesses to integrate these enhanced models into AI agents. The model's widespread availability in over 180+ countries, except the EU, and its new audio (speech) understanding capabilities are expected to facilitate a wide range of applications, from customer service to data analysis.