VFlash 发布：并行输出几何预测，机器人推理提速 5.1 倍
Jim Fan 分享的 VFlash 让机器人直接一口气吐出整条轨迹，不用逐 token 解码，在 Jetson Thor 上推理快了 5.1 倍，精度反而更高，代码还开源了。
VFlash 是一种面向 Physical AI 的 System One 视觉模型思路：物体定位、机器人轨迹规划、稠密 3D 运动等预测任务输出长度固定（如边界框 6 token、10 点机械臂轨迹 54 token），因此可以一次性并行输出整个几何形状，而非逐 token 自回归解码。在 Jetson Thor、DGX Spark 等边缘 GPU 上，VFlash 实现最高 12 倍的动作解码加速和 5.1 倍的整机推理提速，同时 3D 轨迹误差最多降低 61%，原因是避免了自回归解码的级联误差。项目开源。
The era of System One Vision Model is here! Introducing VFlash: parsing the visual world in parallel at unmatched speeds for Physical AI. Most embodied prediction tasks like object localization, robot trajectory planning, and dense 3D motion are fixed-length float vectors. For example, a bounding box is always 6 tokens, a 10-waypoint robot arm trajectory is 54: we know these beforehand. Instead of decoding one token at a time, we can output the entire geometric shape in one go. VFlash enables up to 12x faster action decoding and a 5.1x overall robot inference speedup on edge GPUs like the Jetson Thor and DGX Spark. What’s surprising is that it ends up even more accurate than the VLM backbone, with up to 61% lower 3D trajectory error. The intuition is that VFlash avoids the cascading errors of autoregressive decoding, where every coordinate is written in ink and every later point loses the chance to backspace. Everything that moves will be autonomous, and everything autonomous will need to think on its own chip. The next inference revolution is physical reasoning at the speed of reflex. And as always, it’s open paper, open source! Link in thread: Your browser does not support the video tag. 🔗 View on Twitter 💬 38 🔄 19 ❤️ 174 👀 8677 📊 66 ⚡
