Brain
Figure AI's onboard dual-system VLA. System 2 is an internet-pretrained VLM (7B params, 7-9 Hz) handling scene and language understanding; System 1 is a fast reactive visuomotor policy (80M params, 200 Hz). Helix-02 (January 27, 2026) extended to full-body control via System 0, a 1 kHz neural prior trained on 1,000+ hours of human motion data that replaced roughly 109,504 lines of hand-engineered C++. The first VLA to control a full humanoid upper body including individual fingers from one set of weights, and the first to run two robots from one set of weights.
Foundation model · Maturity: Commercial · Closed · Powers 2 robots
Machine-readable surfaces
- Markdown mirror: /brains/helix.md
- RSS feed: /brains/helix/feed.xml
- JSON-LD: embedded in this page’s head
- REST API: /v1/brains/9fc8ad4b-3d2f-4a3e-a0fe-f7ea7e138eb2
- Data documentation: /data
Architecture
Dual-system VLA. System 2 (S2): onboard internet-pretrained VLM, 7B params, 7-9 Hz. System 1 (S1): fast reactive visuomotor policy, 80M params, 200 Hz. Helix-02 added System 0, a 1 kHz neural prior trained on 1,000+ hours of human motion data, replacing roughly 109,504 lines of hand-engineered C++. Runs on embedded low-power onboard GPUs, no cloud.
Key facts
- Verified commercial milestone
- May 13, 2026: Figure CEO demonstrated a team of Figure 03 robots running a full 8-hour autonomous shift on Helix-02, package sorting, no human intervention
- Training data
- Learned from approximately 500 hours of demonstration data
- Onboard runtime
- Embedded low-power onboard GPUs, no cloud
- DEPLOY maturity rationale
- Verified 8-hour autonomous shift in commercial sortation; counted as commercial not production
Developed by (1)
Powers (2)
Sources (4)
- Figure AI: Helix (primary) · https://www.figure.ai/news/helix
- Figure AI: Helix in logistics · https://www.figure.ai/news/helix-logistics
- TechTimes: Helix-02 8-hour autonomous shift (May 14, 2026) · https://www.techtimes.com/
- Luigi Freda: Helix architecture overview · https://luigifreda.com/
Canonical ID 9fc8ad4b-3d2f-4a3e-a0fe-f7ea7e138eb2