Multimodal perception
Reading images, audio and video as well as text.
capabilityBuilt today Open on the map →
Facts
- Lane
- Models and thresholds
- Wave
- 4 of 12 (build order, not a date)
- Needs
- 7 boxes all the way back through hard links
- Unlocks
- 25 boxes downstream through hard links
- Leverage
- None: already built
- Cluster
- Robots and the physical economy
Evidence as of October 2026
On Karpathy’s list of what agents still lack before they can work like an intern.