Google's Genie 4 generates playable worlds live from a prompt, audio included
prasenx · x · 2026-10-28
Google released Genie 4 as a research preview: a world model that generates an entire playable game live from a single prompt. It steers world, camera and actions in real time, generates native scene-matched audio with talking characters, produces photorealistic scenes with real-world physics, and integrates Gemini reasoning so events unfold step by step. Contrast: Claude can code a whole browser island game, but Genie does it from one prompt — though nobody can ship it yet; Ultra subscribers get access next.
Related event: DeepMind Unveils Genie 4, Its Most Realistic Interactive World Model(41 posts)→
More from Multimodal
- Hands-on with SpAItial Echo-3: record-shop scan turned virtual experience, pricing detailed — risingodegua · 2026-10-28
- T-DIG diversity policies zero-shot transfer to GPT Image 2 and Gemini — simi_97k · 2026-10-28
- Reward hacking in T-DIG: diversity-only reward yields blank images and off-prompt sets — simi_97k · 2026-10-28
- T-DIG releases first text-conditioned diversity dataset with 20k filtered prompts — simi_97k · 2026-10-28
- T-DIG demo: ask for Halloween costumes, get food puns and ML researcher outfits instead of 8 witches — simi_97k · 2026-10-28
- MiniMax brings real-time AI cinema Whispers to Sydney Opera House at NeurIPS — MiniMax_AI · 2026-10-28