What's this about?
A 5B world model streams 832×480 at 24 FPS with joint keyboard/text control and released weights/code; its estimated $0.009 per stream-minute is a server-rental estimate, not a guaranteed retail price (25 HF upvotes).
What exactly is world model?
A model predicts and generates the next visual state under user actions. Playable-looking output does not imply consistent physics or stable long-horizon gameplay.
Why does this matter?
Author-reported preprint; no discussion sampled.
- Significance: text can change events during continued navigation.
- Significance: latency and consistency need real-world tests.
- Significance: rental assumptions shape the cost estimate.