NVIDIA ships Alpamayo 2 Edge: 2B reasoning VLA model running in-vehicle at 160-190ms latency
tomaszbednarz · x · 2026-11-19
NVIDIA released Alpamayo 2 Edge, a 2B-parameter adaptive reasoning vision-language-action model designed for real-time deployment on NVIDIA DRIVE AGX Thor.
- Generates Chain-of-Causation reasoning only when needed; a diffusion action expert produces driving trajectories from surround cameras and navigation inputs
- Closed-loop RL in AlpaGym improves driving performance by 48.9% over the best non-RL checkpoint
- AlpaJudge offers human-calibrated reasoning-quality evaluation (reference-based and reference-free variants), usable as post-training rewards
- TensorRT and DFlash2 optimization achieve 160-190ms pixel-to-trajectory latency on DRIVE AGX Thor
- Model and evaluator available on Hugging Face; deployment recipes in the Alpamayo Developer Hub
More from Embodied
- Robotics sim tips: calibrate your real robot first, never hand-tune sim parameters — jonstephens85 · 2026-11-19
- Embedl cuts Jetson Thor driving-plan latency from 603.7ms to 82.7ms without touching weights — AIFlow_ML · 2026-11-19
- BiCAT builds bimanual coordination directly into VLA action tokens — DJiafei · 2026-11-19
- Game2Player Turns 568 Web Games Into Verifiable Training Grounds for Visual Agents — UIUC-CS · 2026-11-19
- NeoteAI open-sources 5,000 hours of real tactile data plus three tactile foundation reports — 青稞AI · 2026-11-19
- Musk predicts Tesla Optimus will wash, dry and put away dishes in 12-18 months — Polymarket · 2026-11-19