FreeLLMGoogle DeepMind previews Genie 4, a real‑time world model with native audio
Genie 4 creates photorealistic, interactive environments with sound and dialogue, now in a research preview for trusted testers.
Google DeepMind unveiled Genie 4, its latest real‑time world model that renders photorealistic scenes and generates ambient audio, music‑like sound, and spoken dialogue.
Users control the simulation with keyboard inputs and plain‑language prompts that can change weather, move the camera, or add objects without restarting the scene.
The system integrates Gemini Omni for visual grounding and Gemini 3.8 Flash as a planner that interprets prompts, breaks goals into steps, verifies progress, and adapts when the world shifts.
Genie 4 runs on Google‑hosted infrastructure and is currently available as a research preview to trusted testers in creative, journalistic, educational, gaming, and engineering fields.
The preview includes safety guardrails, SynthID watermarks on generated media, and acknowledges limitations such as visual drift, short session length, and imperfect object permanence.
O que isso significa para você
Genie 4 offers developers a new way to prototype interactive simulations with built‑in audio, potentially speeding up pre‑visualization and robotics research. However, as a cloud‑only, early‑stage system it does not yet provide a free or locally runnable alternative, keeping the focus on paid Google AI Ultra and early‑access partners.
No FreeLLMArena
Dos nossos rankings em 31 out 2026, baseados nas votações às cegas da Arena:
- Gemini 3.8 Flash (High): #10 of 419 in Chat, paid
Fontes
- ICYMI: Google DeepMind previews Genie 4 with native audio, TestingCatalog, 28 Oct 2026
- Meet Google DeepMind's Genie 4: Real-Time World Model With Native Audio, MarkTechPost, 28 Oct 2026
Escrito automaticamente a partir das fontes acima, depois verificado pelo nosso código: todo número e nome deve aparecer nelas, e nenhuma frase pode copiar uma fonte. Encontrou um erro? Reporte‑o.
Qual IA você pode usar grátis hoje?
Todo modelo de IA gratuito, classificado por votos humanos cegos, com a memória e velocidade que ele precisa na sua própria GPU.