From RT-1 to MolmoAct2: A Brief History of Vision-Language-Action Models
k7agar · x · 2026-11-02
Robotics community blogger starlitmatcha published a long-form post tracing the architectural evolution of Vision-Language-Action (VLA) models from Google's RT-1 to MolmoAct2, with personal commentary. VLAs are one of the hottest topics in robot learning, often floated as a candidate for a "GPT moment for robotics".
More from Embodied
- China produced 40,000+ humanoid robots by June 2026; 100,000 full-year figure is only a forecast — ingliguori · 2026-11-02
- Dev recreates 'subvocal' for AirPods Pro 3 with one prompt in 9 minutes, open-sourced — _AustinCalvert_ · 2026-11-02
- Scaling past 100 robots without an army of field engineers: 5 ops practices — StewartalsopIII · 2026-11-02
- Neuralink Patient 22 Demonstrates Controlling Robots Telepathically in New Video — Scobleizer · 2026-11-02
- Little grok bot takes its first wobbly steps, then gives up abruptly — Baconbrix · 2026-11-02
- Two UIST 2026 papers in one demo: thermal cold illusion and haptic touch ring — plopesresearch · 2026-11-02