Google DeepMind unveils Genie 4: photoreal real-time world model with physics, native audio and controls
altryne · x · 2026-10-28
Google DeepMind 官方公布了 Genie 4 世界模型的新特性:
- 照片级真实感:基于 Gemini Omni 构建,空间与人物细节逼真
- 高级物理模拟:空气动力学、碰撞检测、自然的人物动作
- 实时控制:通过 prompt 即可改变场景、添加元素或与角色互动
- 原生音频:环境音、音乐、背景噪声与可交互对话
- Gemini 推理:支持多步骤规划
博主 altryne 表示本周将在 ThursdAI 播客中跟进介绍。
Related event: DeepMind Unveils Genie 4, Its Most Realistic Interactive World Model(31 posts)→
More from Multimodal
- Three-Step Halloween Nightmare Effect: Edit, Animate, Then Add Music — miilesus · 2026-10-28
- Three-step Halloween nightmare effect: photo makeover, animation, spooky music — miilesus · 2026-10-28
- DeepMind unveils Genie 4 world model with real-time interactive simulation and native audio — alejandroll10 · 2026-10-28
- Local AI Music Pipeline: LLM Subagents + YuE2 Produce a Full Eurodisco Track — Robonotes1760 · 2026-10-28
- ComfyUI v0.43.0 Ships With Synchronized Graph Cleanups — Fcking_Chuck · 2026-10-28
- Stunning dark etching-style AI illustrations: lurking beasts, starry sails and flower cloaks — OVolosin82152 · 2026-10-28