Google's Genie 4 goes live on Gemini API: steerable 720p worlds via WebRTC
_philschmid · x · 2026-11-20
Google has rolled out Genie 4 on the Gemini API and AI Studio. The world model seeds a 720p 24fps interactive world from text or an image, which you can steer live with WASD keys, prompts, and 48 kHz stereo audio.
Streaming works via WebRTC (browser, with ephemeral auth tokens so the API key never leaves your server) or WebSocket in Python and JS. The sample code shows the full flow: mint a token server-side, connect the client, seed the world with a worldsetup event, then send useraction events for live steering and call complete() to end billing.
More from Models
- Naval-war game benchmark: author claims GPT 6.2 Astra crushes Opus 5.2 — RichardsonDx · 2026-11-20
- Qwen3.8-27B-pi-adapted GGUF quantization trends on Hugging Face — harness-hub · 2026-11-20
- One neuron per language hits 99.1% language-ID accuracy, beating dedicated models — julianweisser · 2026-11-20
- 10 hands-on lessons on what actually happens inside an AI after you type a prompt — techartist_ · 2026-11-20
- Redwood questions Google reusing safety results from weaker model for stronger Carbon — YvesMulkers · 2026-11-20
- GTE-MLM sparse embedding models now natively supported in Transformers v5 — tomaarsen · 2026-11-20