Tag
Wizard 提出一种游戏世界模型,通过解耦角色动画、场景操控与游戏逻辑,实现对未见过角色的关节级动作与实时交互控制,支持 24 FPS 交互并对新场景、角色和动作具备零样本泛化能力。
StreamingSwap is a real-time flow-matching framework for video face swapping that uses a Joint Rec-ID loss and context-efficient generation (Face-RoI + short-context rollout) to achieve 20 FPS on an RTX 4090 over videos up to 10 minutes, along with the HumanFaceVid 1,000-hour cinematic dataset.
Vivix Labs 发布实时交互式视频模型 W1,定价仅 $0.003/秒,价格约为 Seedance 2.5 单条片段的八十分之一,有望让实时交互视频成为常态。
Odyssey 3 是一个基础世界模型,能从提示生成交互式环境并实时预测其变化,其 Pro 版本在 Physics-IQ 视频到视频基准上取得最高分(66.1),还可用于控制机器人手臂、人形机器人、驾驶车辆及训练智能体。
WorldSonus is an interactive video-to-audio framework that brings real-time spatial sound synthesis to world models, using a streaming causal autoregressive diffusion architecture with a real-time factor of 0.41, chunk-indexed prompt scheduling for mid-stream sound control, and stereo supervision for spatial alignment.
The tweet discusses the potential of AI to generate real-time games like Gal Games, drawing parallels to technological revolutions in media from painting to AI.
Meta AI introduces Muse Realtime Avatar, a state-of-the-art embodiment technology that turns voice into synchronized, expressive avatars in real time using a streaming Diffusion Transformer.
PixVerse R2 is a real-time world model that generates continuously evolving audiovisual worlds, accepting text, images, audio, and actions with session memory for interactive and controllable experiences.
Jev can generate game levels in real time, offering faster and cheaper structured output that could significantly impact game development.
A playable demo uses the GPT-Live-1 API to enable real-time world generation in games via voice commands, illustrating a new form of interactive media.
Gemini 3.7 Flash demonstrates the ability to generate interactive websites in real-time based on user input, showcasing its speed and capabilities in AI-driven web development.
LingBot-World 2.0 is an open-source AI model that generates persistent, real-time game worlds at 720p 60fps, running for over 60 minutes without visual decay, using a 14B parameter model with a 1.3B lightweight version for single-GPU deployment.
LingBot-World 2.0 (Infinity) has just been open-sourced, supporting hour-level real-time generation with no quality degradation over long runs and stable 720p/60fps output. It's a key step for robots to learn to predict the world.
A new demo called Reelquest uses Seedance 2.0 and GPT Image 2 to create a fully interactive, playable game where every frame is AI-generated based on user actions, requiring no coding or game engine.
CausalCine is a new academic framework for real-time, interactive multi-shot video generation that uses causal modeling and dynamic memory routing to improve cross-shot coherence in autoregressive models.
Overworld releases Waypoint-1.5, a real-time video world model designed for everyday GPUs, featuring improved visual fidelity and new 360p and 720p tiers for broader hardware accessibility.
Google DeepMind 发布 Genie 4,这是其迄今最真实、最可交互的世界模型,基于 Gemini 构建,支持实时交互、键盘导航与自然语言控制场景,正向可信测试者和 AI Ultra 用户推出研究预览。