The Robotics Stack introduced Real-Time Action Chunking (RTC), an inference-time method from Physical Intelligence @physical_int for running flow and diffusion-based vision-language-action models in real time.
Public source
Publisher name
Public post
NEW METHOD TO TRAIN ROBOTS: Physical Intelligence is at it again! Not so new actually, but worth mentioning because its impact is still going strong today. Called Real-T…
Company
The Robotics Stack
Weekly podcast on what's actually shipping in robotics
- Industry
- Technology, Information and Media
- Company size
- 2–10 employees
About The Robotics Stack
Weekly podcast on what's actually shipping in robotics
See moreLatest activity
Latest activity from The Robotics Stack
15 signals
Customers & Market
The Robotics Stack reported that Unitree's IPO filing shows its average humanoid selling price dropping from about $85,000 in 2023 to about $25,000 last year.
Presence & Recognition
The Robotics Stack announced the The Robotic Origami Challenge, a dexterous-manipulation competition and benchmark held at IROS 2026 organized by 13 co-organizers with the Nippon Origami Association as judge and task curator.
Research & Knowledge
The Robotics Stack notes that Unitree's own filing puts roughly three quarters of its humanoid units into research and education.
Discover more
Similar signals
Similar public activity from other companies.
Research & Knowledge
RobCo
RobCo published an episode of RobTalk featuring Felix Frank discussing how locomotion shifted from hand-engineered optimization to data-driven simulation with GPU parallelization.
Research & Knowledge
Saturday Robotics
Saturday Robotics highlighted Dream2Flow, CHORD, and MolmoMotion as different paths for robot foundation models.
Research & Knowledge
Exylos
Exylos published new samples on Hugging Face for robotics teams to learn from.
Research & Knowledge
Hello Robot Inc
Hello Robot Inc published research from the ETH Zürich Mobile Robotics Lab featuring Stretch robot using Dynamics-Guided Action Correction to convert near-failures into corrective training data.
Research & Knowledge
Robbyant