Google has revealed its plans for the future of AI in enterprises and businesses alike with the release of new upgrades for its Gemini models.
The company has unveiled Gemini 1.5 Pro, promising greater performance alongside, "a breakthrough in long-context understanding" - meaning longer-form prompts could soon be possible.
Alongside this, Google Cloud has also expanded access to Gemini models for Vertex AI customers, meaning more businesses will be able to start using the platform to unlock greater productivity and efficiency for their workers.
Gemini 1.5 Pro
Google only revealed its initial build of Gemini a few weeks ago, but it's clear it is now becoming a major focus as the use of AI soars around the world.
It is set to release an upgrade to all three versions of the platform, but for now only Gemini 1.5 Pro is being rolled out as a preview for developers and enterprise customers.
The company says Gemini 1.5 Pro is able to offer such a boost in performance due to being built on an innovative new version of Mixture-of-Experts (MoE) architecture. Simply put, this brings multiple AI models to work together on different areas to help provide answers, which Google says makes it more efficient to train and service.
Gemini 1.5 Pro will also feature an "experimental" feature concerning long-context understanding, meaning users will be able to submit longer prompts. The regular model has a context window of up to 128,000 tokens (the data required for LLMs such as Gemini to process and generate text) - but Gemini 1.5 Pro will be on offer with up to a million tokens, meaning it can handle more information and provide much richer and more detailed answers.
Google says its step forward now means Gemini 1.5 Pro can process 1 hour of video, 11 hours of audio, codebases with over 30,000 lines of code or over 700,000 words in a single prompt.
The preview of Gemini 1.5 Pro is available now via AI Studio and Vertex AI, but you'll need to be one of a very select group of customers to get it - although we'd expect a wide rollout to launch soon.
"Our teams continue pushing the frontiers of our latest models with safety at the core. They are making rapid progress," noted Alphabet and Google CEO Sundar Pichai in a blog post announcing the news.
"This new generation also delivers a breakthrough in long-context understanding. We’ve been able to significantly increase the amount of information our models can process — running up to 1 million tokens consistently, achieving the longest context window of any large-scale foundation model yet...They will enable entirely new capabilities and help developers build much more useful models and applications."
To continue this rollout, Google Cloud has also announced it is bringing more Gemini models to its customers, with Gemini 1.0 Pro and Gemini 1.0 Ultra now on offer.
This means businesses will be able to utilize the Gemini API within its Vertex AI offering to build the next generation of AI agents and apps.
Google says there are a huge number of possible use cases on offer, not just in terms of summarizing long documents or files, but in analyzing an entire code library in a single prompt, or comparing content across hours of video, and providing "hyper-personalized experiences" for customers coming to your businesses.
