DreamVu's SABER Dataset Delivers 2.19× Performance Gain on NVIDIA GR00T — Captured From Real Stores, Not Simulators

Real-world human capture more than doubles humanoid manipulation success on retail benchmark, outpacing simulation-only fine-tuning of NVIDIA GR00T N1.6 by 2.19×.

PHILADELPHIA, PA — May 12, 2026 — DreamVu today released SABER, a high-fidelity robotics dataset built from 100+ hours of natural in-store human activity, along with research demonstrating that post-training NVIDIA's GR00T N1.6 vision-language-action model on SABER more than doubles its success rate on retail manipulation tasks — without a single robot in the loop.

Out of the box, GR00T N1.6 — NVIDIA's flagship humanoid foundation model — scores near zero on RoboBenchMart, a standardized retail manipulation benchmark. With SABER applied as a domain-specific post-training layer, the same model reaches 29.3% mean success across ten tasks, a 2.19× improvement over simulation-only fine-tuning (13.4%). Fridge interaction tasks reach 82–100% success.

"Synthetic data has been the industry's default path to scale in humanoid training, and SABER shows the ceiling of that approach. A humanoid robot doesn't deploy into a render — it deploys into a real store, with real packaging, real lighting, and the messy repetition of real human behavior."
— Rajat Aggarwal, CTO and Co-Founder of DreamVu

Dataset

SABER is built from DreamVu's synchronized ego + exo capture system. Head-mounted cameras worn by primary actors record fine-grained hand activity at the point of interaction, while Alia360 simultaneously observes the full scene from a single fixed unit. The resulting corpus contains approximately 44.8K training samples across three complementary streams: 25K latent action sequences from egocentric video, 18.6K dexterous hand-pose trajectories retargeted to robot joint space, and 1.2K whole-body motion sequences retargeted to the Unitree G1 humanoid embodiment.

Paper

The SABER paper is at arXiv.

About DreamVu

DreamVu designs and builds patented 360° 3D cameras for robots, and the software that runs on them. The company is headquartered in Philadelphia, PA, with offices in Palo Alto, CA and R&D in Hyderabad, India.