Lumos Robotics' industrial embodied brain Prime R0 ranked first worldwide by overall success rate across all single-arm fine manipulation and dual-arm collaborative manipulation tasks on MolmoSpaces, a zero-shot embodied intelligence benchmark initiated by AI2.
At the inaugural EgoCross Challenge of the CVPR 2026 EgoVis Workshop, Nuoyin Intelligence took first place globally in both the Source-Limited and Open-Source tracks, becoming the only team to win championships in both tracks.
Aleph Surgical released a research preview of π0.5 driving autonomous surgical anastomosis procedures.
Teams including Berkeley AUTOLAB propose CaP-X, an open framework for systematically studying Code-as-Policy agents in robot manipulation, centered on the interactive environment CaP-Gym and CaP-Bench, which evaluates 12 frontier models.
Why it matters: The original text presents a systematic evaluation framework for Code-as-Policy agents and the finding that test-time scaling can mitigate abstraction dependence, offering reference value for research on coding-agent manipulation control.
Dexterity published an article introducing system evaluation and production measurement for the Foresight packing agent. On the planning side, it was tested on four real warehouse box-type distributions, with density of 0.65–0.77 and stability of 99.7–100%, and distribution-free order statistics were used to give 95% confidence intervals; the article also found that randomly shuffling the box sequence causes local variance to collapse by about 1.4 times, making evaluation overly optimistic.
Why it matters: The original article fully discloses the evaluation method for the packing agent, including distribution-free confidence intervals and the evaluation bias caused by random shuffling; the method can be transferred to similar systems.
The Toyota Research Institute (TRI) published a tutorial article introducing a statistical A/B testing method based on success/failure labels for robot policy evaluation.
Why it matters: The TRI authors systematically explain statistical pitfalls and reusable tools in robot policy A/B testing, with runnable Python examples.
In a blog post, Physical Intelligence argues that robots need a reusable physical intelligence layer, and it published real-world deployment data from partners Weave and Ultra using π 0.6.
Why it matters: The two deployers provided comparative data on autonomy rates, missed grasps, and throughput, making it easier to judge the practical gains that foundation model iterations bring in real-world settings.
The Kepler K2 "Bumblebee" humanoid robot completed three rounds of external push disturbance tests on a slope, staying stable in all of them, and said, "Push all you want — I'll keep walking." This is a real-machine stability test; the company says continuous training is still ongoing, and more real-machine performance results will be released gradually.
By fine-tuning its latest π 0.6 model, Physical Intelligence achieved preliminary solutions for the gold-medal tasks in 3 of the five everyday manipulation challenges in the Robot Olympics proposed by Benjie Holson, and reached silver-medal level in the remaining 2, with an average success rate of 52% and task progress of 72%.
Why it matters: Using the same set of externally proposed everyday manipulation tasks to compare performance with and without robot pretraining lets readers judge the practical value of foundation-model pretraining for manipulation skills.
Livox has released a performance test report for its automotive-grade hybrid solid-state LiDAR HAP; the full text is available on request. In testing, HAP effectively detected a white sedan at 150 meters, still returned more than 7 points on the vehicle body at 200 meters, and detected more than 3 points on a 15-centimeter water bottle at about 30 meters.
Chuangxin Hulian's medical service robot "Xiaoxin" (the name can be customized) is designed for hospital inquiry and guidance scenarios. The robot has a humanoid design, stands about 1.5 meters tall, and consists of three parts: head, body, and base. The test unit uses an ABS engineering plastic shell, and the manufacturer says an aluminum alloy shell can be customized on request; its "face" features an 8-inch LCD screen that displays virtual expressions to bring the human-machine distance closer.