Google has recently pulled back the curtain on a preview of its updated large language model, Google Gemini 1.5. The tech giant’s latest AI breakthrough aims to deliver a more efficient “long-context understanding,” meaning it can interpret and relate to a larger scope of information than ever before.
One of Gemini 1.5’s standout features is its substantially expanded potential context window. This AI model can handle up to 1 million tokens, a significant leap from its predecessor’s 32,000 tokens and far beyond the competitors’ 128,000 tokens. Tokens are units of data the AI uses to understand context, so the larger the token window, the more complex and nuanced the data the AI can process.
Thanks to this expanded token context window, Gemini 1.5 can interpret up to one hour of video, 11 hours of audio, over 700,000 words, or over 30,000 lines of code. This makes it an incredibly versatile tool for processing and understanding a wide range of data types.
To begin with, early access to Gemini 1.5 and its full token window is granted only to a select group of developers and business customers. These individuals and organizations can access the AI via Google’s AI Studio and Vertex AI dev platforms.
When officially released, the entry-level tier of Google’s Gemini 1.5 Pro will start with a 128,000 token window. Pricing will scale upwards depending on the size of the token window required, up to the maximum 1 million token limit.
The unveiling of Gemini 1.5 comes as Google’s response to its competitor, OpenAI, which recently incorporated persistent memory into its ChatGPT. Google’s advanced and compact Gemini model was developed by a researcher at Google DeepMind, the company’s AI research lab.
Currently, Google’s most powerful model on offer is Gemini Ultra 1.0. For a fee of $20 per month, this model can assist users with a variety of tasks, from changing tires to analyzing financial statements.
The upcoming Gemini 1.5 Pro model is primed to offer similar performance to the Ultra 1.0 model. However, it will be less expensive and smaller, making it a more appealing choice for corporate customers and AI researchers. Access to this model will be via the Gemini API on Google’s AI Studio sandbox and Google Cloud, giving users a variety of access points to this powerful AI tool.
Google continues to push the boundaries of AI with its Gemini 1.5 model, promising enhanced understanding of complex data and superior performance for users across a variety of applications.
Connect with our expert to explore the capabilities of our latest addition, AI4Mind Chatbot. It’s transforming the social media landscape, creating fresh possibilities for businesses to engage in real-time, meaningful conversations with their audience.
