IEEE Spectrum rounds up robotics videos, including a Skydio fixed-wing drone using a robot arm for launch and capture, humanoid performances, and underwater manipulation. The roundup does not establish autonomous operation for the humanoid performance.
Editorial context：A robotics video roundup. Demonstrations do not establish deployment readiness or autonomous operation.
Animesh Garg@animesh_gargSignalEditorial score8585Introducing FLUX 3 Action. An open weights 7B World Action Model that achieves first place on the RoboLab benchmark. It outperforms the previous best open model by 6.1 percentage points while using 56% fewer parameters and running up to 3.95x faster. FLUX 3 Action removes the usual trade-off between world action model performance and VLA speed: it still predicts video and actions together, but plans more than twice as far ahead and runs faster per second of robot motion than the strongest open VLA. Teams can fine-tune FLUX 3 Action on their own demonstrations to create policies for a particular robot and task. Together with @nvidia, we also integrated FLUX 3 Action natively into @huggingface's LeRobot, with fine-tuning recipes included and edge deployment on NVIDIA Jetson. Beyond robotics, we’re also seeing promising results training task-specific policies for acting in simulated environments like gaming, controlling a vehicle, computer use, and wherever else a model needs to understand a visual environment and then choose what to do next. FLUX 3 Action builds on the same image, video, and audio pretraining as FLUX 3, but uses a smaller architecture designed for practical deployment. In midtraining, we trained the model to predict actions and future frames together. We’re releasing the weights, code, fine-tuning recipe, benchmarks, and reproducible examples so researchers and developers can build on the model with their own robots, environments, and tasks (see below).
Editorial context：FLUX 3 Action is a 7B parameter world action model that achieves state-of-the-art performance on the RoboLab benchmark, outperforming previous models with fewer parameters and faster inference. It integrates action prediction with video generation and supports fine-tuning for specific robots and tasks.
NVIDIA and Hugging Face are releasing the NVIDIA Isaac GR00T 1.7 model and Isaac Teleop framework into LeRobot, an open-source robotics library, to provide developers with shared tools for training, evaluating, and deploying robot foundation models. NVIDIA Cosmos 3, a frontier world model for physical AI, is also set to be integrated soon.
Editorial context：NVIDIA and Hugging Face are collaborating to integrate advanced models and frameworks into LeRobot, an open-source robotics library, to streamline end-to-end robot development and foster community innovation.
This tutorial explores the challenges of deploying VLA models on embedded robotic systems, including dataset recording best practices, fine-tuning techniques for ACT and SmolVLA, and real-time performance optimization using the NXP i.MX 95 SoC. It emphasizes asynchronous inference and hardware-aware scheduling to improve control and reduce latency.
Editorial context：This guide provides hands-on best practices for deploying Vision-Language-Action (VLA) models on embedded platforms, emphasizing dataset recording, model fine-tuning, and real-time performance optimization. It highlights the importance of asynchronous inference and hardware-specific optimizations for achieving reliable robotic control.
LeRobot v0.4.2 includes bug fixes, performance improvements, and new features like real-time chunking and dataset encoding optimizations. Updates address episode filtering, script porting, and environment preprocessing.
This hands-on tutorial walks through the process of collecting data, training policies, and deploying autonomous medical robotics workflows on real hardware using NVIDIA Isaac for Healthcare. It introduces the SO-ARM starter workflow, which enables developers to build and validate surgical assistant robots from simulation to deployment.
Editorial context：This tutorial provides a comprehensive guide to building a healthcare robot using NVIDIA Isaac for Healthcare, covering data collection, simulation, training, and deployment on real hardware. It emphasizes the use of simulation to generate synthetic data and the integration of real-world data for training policies that generalize across domains.
Hugging Face announces LeRobot v0.4.0, a major upgrade for open-source robotics with Dataset v3.0, new VLA models like PI0.5 and GR00T N1.5, and a plugin system for hardware integration. The release also adds support for LIBERO and Meta-World simulations, multi-GPU training, and a new Hugging Face Robot Learning Course.
Editorial context：LeRobot v0.4.0 introduces significant upgrades for open-source robotics, including Dataset v3.0 with chunked episodes and streaming capabilities, new VLA models like PI0.5 and GR00T N1.5, and a plugin system for hardware integration. These enhancements improve scalability, data management, and support for simulation environments like LIBERO and Meta-World.
Open Robotics reports record nine students participating in GSoC 2025 across five projects. Key projects include ROS 2 tracing improvements, C/C++ to ROS message conversion, and Gazebo-based sonar simulations.
NVIDIA has released the GR00T N1.5 model, a cross-embodiment foundation model for generalized humanoid robot reasoning and skills. The model can be fine-tuned using teleoperation data from a SO-101 arm, with a detailed tutorial provided for developers. The release includes instructions for dataset preparation, fine-tuning, evaluation, and deployment.
Editorial context：NVIDIA's GR00T N1.5 is a cross-embodiment model for generalized humanoid robot reasoning and skills, adaptable through post-training for specific tasks and environments. The release includes a step-by-step tutorial for fine-tuning using teleoperation data from a SO-101 arm, emphasizing the use of the EmbodimentTag system for customization.
Hugging Face introduces SmolVLA, a 450M parameter open-source Vision-Language-Action model for robotics, trained on community-shared datasets. It outperforms larger models in simulation and real-world tasks, supports asynchronous inference for faster response, and is designed for deployment on consumer hardware.
Editorial context：SmolVLA demonstrates that compact, open-source models can outperform larger proprietary systems in both simulation and real-world tasks, highlighting the potential of community-driven data and efficient architectures in advancing robotics research.
