NVIDIA releases Alpamayo 2 Edge autonomous-driving model
NVIDIA released Alpamayo 2 Edge on November 10, an end-to-end autonomous-driving VLA that combines causal reasoning, adaptive inference, and a diffusion action expert. The approximately 2.45B-parameter VLM backbone and 0.2B-parameter action expert run on DRIVE AGX Thor, with NVIDIA reporting a 48.9% closed-loop driving gain after reinforcement learning and a 160–190 ms Pixel-to-Trajectory latency in its reference configuration. The performance figures are vendor-reported and primarily reflect simulation-based evaluation.
Alpamayo 2 Edge provides a detailed automotive VLA approach that links explainable causal reasoning with continuous trajectory generation while managing onboard inference latency.
导语:她叫Alpamayo 2 Edge,因果推理结合扩散动作,端到端输出闭环表现提升 48.9%。

她叫Alpamayo 2 Edge,因果推理结合扩散动作,端到端输出闭环表现提升 48.9%。
作者丨郑佳美
编辑丨岑 峰
点开前,你以为老黄家的惊天大瓜又有了什么后续; 点开后才发现:这个“她”连人都不是。
#NVIDIA#Alpamayo 2 Edge#DRIVE AGX Thor#Alpamayo 2 Super#AlpaJudge#AlpaGym#AlpaSim#TensorRT