News · 9 articles
Deep dives
Technical notes from training and evaluating on our data: π0.5, GR00T, and more.
Deep Dive
Quick Note: π0.5 Inference in 16 GB
Two fixes cut our π0.5 rollout stack from about 22 GB to 11.2 GB of VRAM, so it now fits 16 GB cards like the RTX 5080 and 5070 Ti.
Deep Dive
Latency Versus π0.5 Rollout Performance
A Python pipe buffer added hundreds of milliseconds to our π0.5 rollouts. How we found it, cut inference from 230 ms to 87 ms, and what changed on the robot.
Deep Dive
openpi / π0.5 Fine-Tuning: How-To and Tips
Ablations from a π0.5 fine-tune on YAM pick-and-place data: training scheme, lighting, inference bugs, camera placement and field of view, and backgrounds.
Deep Dive
Intro to AI for Robotics
An introduction to AI-native robotics, covering learning loops, teleoperation data, and how modern robots move from demos to real work.
Deep Dive
Teleoperation: Now and the Future of AI Robotics
Read this article if you’re interested in contributing to the robotics revolution, which will reshape the way we work and interact with the physical world.
Deep Dive
GR00T Part 4: Advancing Robotics Foundation Models
In this final part, we explore how robotics foundation models can move from GPT-2 to GPT-3 era using synthetic, simulator, and high-quality data.
Deep Dive
GR00T Part 3: Predictive Control in AI Robotics
Exploring how predictive models and latent actions help robots learn tasks—from walking to making sandwiches—paving the way for scalable AI robotics.
Deep Dive
GR00T Part 2: Labels Without Labels
Unsupervised training teaches AI to extract powerful features from video, linking compression to real-world actions and advancing robotics learning.
Deep Dive
GR00T Part 1: Foundation Models & the History of AI Robotics
Nvidia’s GR00T-1 marks a GPT2 moment for robotics—proof that foundation-scale models trained on video can generalize across tasks and hardware.