Genie 4

Google DeepMind announced the interactive world model 'Genie 4' and began offering a research preview to trusted testers.

• •Google DeepMind announced the interactive world model 'Genie 4' on October 27.
• •A research preview is currently being provided to trusted testers.
• •It will later be rolled out to Project Genie's Google AI Ultra subscribers and early access partners for Gemini API and Google AI Studio.

Google DeepMind announced 'Genie 4,' an AI that generates interactive virtual worlds in real time, enabling simultaneous audio and conversation generation with Gemini integration.

• •Improved video quality and physics simulation allow real-time simultaneous generation of environmental sounds and conversation audio.
• •It enables natural character movements and environment manipulation via text.
• •Initially offered as a research preview, it will be rolled out to 'Google AI Ultra' subscribers and others.

Google DeepMind released Genie 4, a world model featuring real-time audio, more realistic simulations, and mid-simulation text prompts.

• •Genie 4 supports real-time audio and more realistic simulation environments.
• •It allows users to input text prompts to steer the simulation while it is running.
• •The model is available as a research preview for trusted testers.

Genie 4は、Geminiモデルを駆使したリアルタイムの対話型ワールドシミュレーションやネイティブ音声、高度な物理演算などの機能を備えている

• •ネイティブ音声と高度な物理演算、自然な人間の動きをサポートする機能を実装している。
• •シミュレーションの制御にはGeminiモデルが活用されている。
• •リアルタイムで対話できるワールドシミュレーション環境を提供する。