Google Launches Gemini 3.1 Ultra With 2M Token Context

1 min read

Contents show

Google’s Biggest AI Model Yet

Google has launched Gemini 3.1 Ultra, featuring an unprecedented 2-million token context window for multimodal AI.

The model works natively across text, image, audio, and video – making it truly multimodal from the ground up.

A new Code Execution tool lets the model write, run, and test code mid-conversation inside a secure sandbox.

Gemini 3.1 Ultra competes directly with Claude Mythos 5 and GPT-5.4 in the race for AI supremacy.

The massive context window means developers can feed entire codebases, long documents, and hours of video in a single prompt.

Gemini 3.1 Ultra is rolling out to Google Cloud and AI Studio users starting this week.

The AI arms race is accelerating. Larger context windows and multimodal capabilities are becoming the new standard.