Google DeepMind Releases Genie 4 World Model With Native Audio and Real-Time Steering
Google DeepMind released Genie 4, its newest world model, in research preview for trusted testers. The model generates photorealistic, interactive simulated worlds with native real-time audio, including ambient sound, music and character dialogue, and supports mid-simulation text prompts to change weather, camera angles or add people. It maintains scene consistency for up to two minutes, double Genie 3, with improved physics. Gemini acts as an in-loop director planning steps and checking frames. Access will expand to Google AI Ultra subscribers and early Gemini API partners.
IllustrationEditorial responsibility
- No named human review is recorded for this page.
- Reports are grouped by semantic similarity and deterministic rules. Language models may assist titles, summaries, translation and cross-source analysis; the page reads the event directly, while its address stays stable when the title changes.
- Summary covers the current reports
Cross-source coverage
Reporting timeline
Google DeepMind Unveils Genie 4 Real-Time World Model With Native Audio
Google DeepMind announced Genie 4, described as its first real-time world model with native audio that generates ambient sound, music, and live dialogue from characters reacting to users and in-world events. Users can rewrite a running simulation by typing, changing weather, moving the camera, or adding people, with Gemini fitting each edit to the scene and keeping it coherent. Gemini 3.8 Flash acts as an in-loop director, splitting goals into steps and checking incoming frames, and Gemini 4 Argon used that loop to simulate Apptronik's Apollo humanoid playing checkers. Access begins with trusted testers before expanding to Google AI Ultra subscribers and early Gemini API partners.
Read sourceGoogle DeepMind unveils Genie 4 AI model for real-time interactive 3D worlds
Google DeepMind announced Genie 4, a new AI model that generates photorealistic, interactive 3D worlds in real time. According to the announcement, the model produces environments complete with physics, sound, and human dialogue. The post presents the release as a just-in development without additional technical details, benchmarks, availability timelines, or pricing. No independent verification or further specifications were included in the source item.
Read sourceGoogle DeepMind Releases Genie 4 World Model With Real-Time Audio for Trusted Testers
Google DeepMind released Genie 4, a world model, in research preview for trusted testers. According to the announcement, the new model adds real-time audio, more realistic simulations, and mid-simulation text prompts. The release is limited to a research preview available to trusted testers rather than a general public launch. No further details on pricing, availability timeline, or technical specifications were provided in the source item.
Read sourceShow 6 older updatesHide older updates
Google DeepMind Releases Genie 4 World Model With Audio and Real-Time Steering
Google DeepMind released Genie 4, its newest world model, in research preview for trusted testers on Tuesday. The model generates more photorealistic and interactive simulated worlds, supports mid-simulation text prompts to change scenes, and produces real-time audio including music, ambient noise, and interactive dialogue. It maintains scene consistency for up to two minutes, double its predecessor Genie 3, with improved physics and more human-like characters. Gemini is more closely integrated, acting as a director that plans steps and checks frames, and a Gemini 4 Argon experiment with Gemini Robotics broke tasks into steps Genie 4 simulated in real time. Research director Tejas Kulkarni described future hybrid world models combining Gemini code and text with Genie simulation. Trusted testers point to film prototyping and storyboarding as a possible use, per engineer Diego Rivas, who stressed it is early days and noted gaps such as detail drift and audio accuracy. Genie 4 will roll out to Google AI Ultra subscribers in Project Genie and early Gemini API and AI Studio partners.
Read sourceGoogle Announces Genie 4 World Model With Native Audio and Dynamic Prompting
Google has officially announced Genie 4, described as its latest World Model featuring advanced realism and interactivity. According to the announcement, users can now dynamically prompt changes to the world and generate worlds with native audio. The model is rolling out to Trusted Testers as a research preview and is slated to reach Google AI Ultra subscribers and early access partners through the Gemini API and AI Studio. The post includes a sample attributed to Google CEO Sundar Pichai. No pricing, benchmark results, or release dates beyond the staged rollout were disclosed in the source item.
Read sourceGoogle Unveils Genie 4 Interactive World Model for Trusted Testers
Google announced Genie 4, described as its most realistic and interactive world model, in a post shared by Sundar Pichai. According to the announcement, the model can steer dynamic environments in real time, shaping events, camera angles, and actions; generate native audio matched to the world, including realistic background sounds and characters that can converse in real time; simulate photorealistic scenes and real-world physics; and create more coherent multi-step events through reasoning from Gemini. Google said it has begun rolling out Genie 4 as a research preview to trusted testers. The company added that it is coming soon to Google AI Ultra subscribers in Project Genie and to early access partners via the Gemini API and AI Studio.
Read sourceGoogle DeepMind launches Genie 4 AI for real-time interactive world generation
Google DeepMind has launched Genie 4, an AI system that generates realistic, interactive worlds in real time, according to a post on X. The post says users can explore a generated world, change it, and watch it react to their actions. Genie 4 is currently in research preview, with broader access described as coming soon. The post frames the release as significant for the future of gaming. No technical details, benchmarks, pricing, or independent verification were provided in the source item, and the launch date was not specified beyond the post's framing.
Read sourceGoogle DeepMind Unveils Genie 4 Interactive World Model
Google DeepMind announced Genie 4, describing it as its most realistic and interactive world model to date. According to the post, users can explore dynamic environments in real time to observe how a scene responds and sounds, and how the people within it move, react, and speak. The announcement positions Genie 4 as an advance in interactive world modeling, emphasizing real-time exploration and responsive, human-like behavior within generated scenes. No technical specifications, benchmarks, availability details, or pricing were disclosed in the post.
Read sourceGoogle DeepMind Releases Genie 4 Real-Time Interactive World Model With Native Audio
Google DeepMind announced Genie 4, a new world model focused on real-time interactive simulation and improved photorealistic quality. According to the company, Genie 4 is the first real-time world model to natively generate audio, including ambient sound and real-time interactive character dialogue. Google DeepMind provided a full official explanation of Genie 4's real-time interaction, native audio, and collaboration mechanisms with Gemini. The company also acknowledged current limitations in long-horizon consistency and physical accuracy.