An open vision-language-action (VLA) foundation model for generalized humanoid robot skills, using language and image inputs for manipulation, with cross-embodiment post-training and humanoid whole-body control support.
Primary use case: Generalized humanoid manipulation and whole-body control, including adaptation to specific robot embodiments, tasks, and environments.
AI Capabilities
Vision-language-action control, Multimodal language and image understanding, Generalized humanoid robot manipulation, Cross-embodiment adaptation and post-training, Human-video pretraining for manipulation-prior transfer, Language-conditioned whole-body humanoid control, Zero-shot inference, Fine-tuning with custom robot data
Onboard Compute
NVIDIA RTX PRO 6000 Blackwell Server Edition or Jetson AGX Thor; also supported are RTX 4090, L40, H100, Jetson AGX Orin, and DGX Spark.
Availability
General Availability (GA); downloadable model weights and reference code are offered through Hugging Face and GitHub, with commercial support for production deployment.
Primary Use Case
Generalized humanoid manipulation and whole-body control, including adaptation to specific robot embodiments, tasks, and environments.
Shareholders approved the Agility Robotics merger on November 2. Closing is expected November 4, followed by Nasdaq trading under AGLT on November 5.
Drug companies demand speedier launches and don’t want to relax quality standards. That puts pressure on the shop floor as well as on the software above it. Rockwell Automation’s October 1 report found that 90 percent of life sciences manufacturers call digital transformation necessary. Nearly 58 percent are already using smart manufacturing technology in their […]
Bedrock Robotics automates large excavators to improve construction site economics. | Credit: Bedrock Robotics San Francisco-based Bedrock Robotics has raised $600 million in Series C funding, co-led by CapitalG and Emergence Capital, at a $4.1 billion valuation. The round brings its total funding to nearly $1 billion