NVIDIA Releases Open-Source Self-Driving Model Alpamayo 2 Edge, 10x Faster on AGX Thor
Course_Latter · reddit · 2026-11-12
NVIDIA released Alpamayo 2 Edge, an open-source self-driving model trained in an RL loop that takes six cameras plus recent motion as input. Embedl, which contributed an optimized version, reports it runs more than 10x faster than PyTorch on edge hardware like the Jetson AGX Thor, with a live demo running on a real Thor.
More from Embodied
- Berkeley open-sources RoboSimStudio: AI agents program industrial robots without training data — berkeley_ai · 2026-11-12
- NXP ships pre-tested drone platform: initial flight in 30 minutes out of the box — shashib · 2026-11-12
- Open-source dual-arm robot solves tangram puzzles with 90 demos, ~60% success — DanGrover · 2026-11-12
- Robotics has three regional playbooks: China volume, Europe inspection, US pilots — ingliguori · 2026-11-12
- 3,500 Laser-Armed AI Submarine Robots Patrol Norwegian Salmon Farms — philfung · 2026-11-12
- Model Autonomously Designs a Surgical-Robot-Style Robotic Arm CAD After 75 Minutes — _Stocko_ · 2026-11-12