Figure Introduces Helix AI Model for Advanced Humanoid Robot Capabilities

Helix enables natural language control, object generalization, and multi-robot collaboration for household tasks.

Helix is a Vision-Language-Action (VLA) model that integrates perception, language understanding, and control to enhance humanoid robot functionality.
The system allows robots to understand natural language prompts, generalize tasks for unseen household objects, and perform complex manipulations without extensive programming.
Helix supports multi-robot collaboration, enabling two robots to work together on shared tasks such as passing and placing objects.
The model operates on low-power GPUs, making it suitable for real-world deployment in household environments.
Figure developed Helix entirely in-house after separating from OpenAI, positioning it as a significant step forward in scaling humanoid robots for home use.