From the course: AI Trends

Unlock this course with a free trial

Join today to access over 25,200 courses taught by industry experts.

New Google Gemini Models and Google I/O Announcements

New Google Gemini Models and Google I/O Announcements

From the course: AI Trends

Start my 1-month free trial Buy for my team

New Google Gemini Models and Google I/O Announcements

“

- Google IO brought a number of features focused on an expansion to Google Gemini. Let's start with new features added to Google's chat bot, Gemini Advanced. Google Gemini Pro 1.5, which used to be only available to developers, is now inside their Google advanced platform. That lets you use a model with more capabilities as well as a huge 1 million token context window. You can use Google Docs, PDFs, and Word files from Google Drive or use your own file uploads. I can easily add a large volume of an old encyclopedia and ask a very specific question from that huge context. That million token context window means you can look through an hour's worth of video, 11 hours of audio, 30,000 lines of code, or a 700,000 word document. Another huge announcement is a low latency multimodal model called Gemini Flash. It offers advanced reasoning and huge context windows of up to a million tokens. You can see on this chart that it performs…

Contents