Figure’s humanoid robot takes voice orders to help around the house
Figure has revealed a Vision-Language-Action (VLA) model for humanoid robots. VLAs leverage vision and language commands to process information, enabling robots to be trained through a combination of video and large language models. Figure's Helix model enables robots to follow natural language commands and pick up novel items with varying shapes, sizes, colors, and material properties never encountered before in training. Work on Helix is still at a very early stage, so demonstrations should be taken with a grain of salt.