DeepMind publishes Genie 4 model card: real-time 720p 24fps world generation with audio, on the API
lyraxana · x · 2026-11-20
Google DeepMind published the Genie 4 model card, confirming the world generation system is available via the API sooner than expected.
Key points:
- Real-time 720p interactive video at up to 24fps with synchronized audio (music, ambient sound, interactive dialogue)
- Built on Gemini Omni Flash, with Gemini handling reasoning and real-time steering; Gemini Omni anchors simulations in real-world appearance and maintains coherence
- Transformer architecture with native multimodal inputs: text, images, and real-time user actions (keyboard/mouse)
- Trained on video, audio, image, and text data with deduplication and safety filtering
Related event: Google DeepMind launches Genie 4 world model as Project Genie goes live(14 posts)→
More from Models
- Meta's Muse AI, 10M+ downloads, hit by new privacy breaches including bank balances — karmay007 · 2026-11-20
- Datalab shows ParseBench is gameable: a 650M model hit near the top with a few RL steps — VikParuchuri · 2026-11-20
- GPT-6.2 Astra matches predecessor's MazeBench score at $225 vs $840 — patience_cave · 2026-11-20
- MazeBench: GPT-6.2 Sol narrows gap with Astra but still scores less at same price — patience_cave · 2026-11-20
- Gemini 4 Neon leads MazeBench on uncached token efficiency by a wide margin — patience_cave · 2026-11-20
- MazeBench: GPT-6.2 Astra tops at 16% while costing 25% of its predecessor — patience_cave · 2026-11-20