MultimodalJake Bruce et al. (Google DeepMind)Google DeepMind

Genie published

Genie published — 11 billion parameters; trained only on unlabeled internet videos, it generates virtual environments that can be controlled without action labels

Why it matters

The starting point of the world-model lineage. The authors argue that at this scale it can be regarded as a foundation world model

Source — Primary sources first; secondary reports are labelled “Press”

Paperarxiv.orghttps://arxiv.org/abs/2402.15391

Added in Week 35, 2026 · See that week

Same week

RLOOPaperGemma 1 (2B/7B), under Gemma's own terms of usePressYang Zhilin frames AI as scaling laws and long context as key to AGIPress

Nearby on the same axis

OpenAI announces Sora with a technical reportOfficialxAI previews Grok-1.5VOfficialGemini 1.0OfficialGPT-4oOfficial