Genie 4 generates editable video and audio in real time

CodeByPoonam · x · 2026-10-28

Genie 4's key capability is real-time video and audio generation: users can steer the output with keyboard commands and prompts while it plays. The quoted DotCSV tweet notes this fourth iteration looks more cinematic than the previous version.

Related event: Google launches Genie 4 real-time world model: generate and direct interactive worlds from a prompt(15 posts)→

Original post →

More from Multimodal

• One AI model, one whole movie: creator shares free multi-stage workflow — mickmumpitz · 2026-10-28
• Camera Thief: film a camera move on your phone, apply it to any still photo (open source) — kevinivarsson · 2026-10-28
• AssemblyAI launches first TTS model: 80ms TTFB, 4,000+ voices, $15/1M chars — ycombinator · 2026-10-28
• VEED launches Dubbing API with emotion-preserving lip-synced multilingual video translation — Med1_Ai · 2026-10-28
• Creator publishes exact prompts behind AI games and videos: 7,200 code-drawn frames and a builder-critic loop — prasenx · 2026-10-28
• Oracle launches image and video AI models trained on rights-cleared data — YvesMulkers · 2026-10-28

Multimodal channel →