Google has introduced new updates to its AI services, including Gemini 1.5 Flash, improvements to Gemini 1.5 Pro, and advancements in Project Astra for future AI assistants. Gemini 1.5 Flash is a quicker, more effective model that can handle large-scale tasks. Although lighter than the 1.5 Pro, it still allows for reasoning across a lot of data types and boasts a groundbreaking long context window of one million tokens.
Demis Hassabis, CEO of Google DeepMind, explained that 1.5 Flash is great at summarizing, chat apps, image and video captions, and extracting data from long docs and tables. It learns from 1.5 Pro through ‘distillation,’ focusing on key info. Google enhanced Gemini 1.5 Pro with a larger context window and better code gen, reasoning, conversation, and audio/visual understanding.
We’ve always wanted to build a universal agent that will be useful in everyday life. Project Astra, shows multimodal understanding and real-time conversational capabilities,
“With technology like this, it’s easy to envision a future where people could have an expert AI assistant by their side, through a phone or glasses.”
The company added Gemini 1.5 Pro to Google products like Gemini Advanced and Workspace apps. Gemini Nano can now recognize multimodal inputs, not just text but images too.
Google introduced new advanced models called Gemma 2 for better performance. They also added PaliGemma, their first vision-language model. Google also updated on Project Astra, a vision for AI assistants of the future, with prototype agents that can understand better and respond quickly in chat.
