Google Gemini 2.0 Released: What's New and What Changed

Google's release of Gemini 2.0 represents a significant step forward in AI capabilities, particularly in multimodal understanding and real-time processing. Here's what's actually new.

Key New Capabilities

Real-Time Multimodal Understanding

Gemini 2.0's most impressive new feature is its ability to understand live video streams. You can point your camera at something and have a real-time conversation about it — useful for accessibility applications, educational tools, and hands-free assistance.

Native Tool Use

Gemini 2.0 natively understands and uses tools: code execution, web search, and third-party APIs can be invoked seamlessly within a single response. This makes building agentic applications significantly simpler.

Reasoning Improvements

Benchmark performance on complex reasoning tasks has improved substantially, with particular gains in mathematics, science, and multi-step planning problems.

Product Availability

• Gemini Ultra: Gemini 2.0 Ultra is available to Google One AI Premium subscribers
• Gemini Pro: Available via Google AI Studio and Vertex AI
• Gemini Flash: Faster, more affordable model for high-volume applications
• Gemini Nano: On-device model for Android Pixel phones

Google Workspace Integration

Gemini 2.0 is now embedded across Google Workspace: deeper integration in Gmail for email drafting and summarization, real-time Google Meet transcription and action items, Sheets data analysis improvements, and Docs writing assistance that understands document context more deeply.

Competitive Position

With Gemini 2.0, Google has significantly closed the gap with GPT-4o and Claude 3.5. The differentiating factors remain Google's unique advantages: real-time web integration, Google Workspace depth, and on-device capabilities through Android.