Jim Fan 团队开源 Flora 机器人数据引擎模型套件
We built a robotics data engine with Astra-level performance at 125x lower cost. And we are open-sourcing the weights today! Introducing Flora, a suite of open
Jim Fan 团队开源 Flora，一套用于解析物理世界、把原始视频转成训练信号的开放模型，称其性能达到 Astra 级而成本低 125 倍。Flora 已支撑其内部数据引擎，在 500 块 GPU 上每天标注 20 万小时视频。它提供带时间戳的高质量描述用于筛选人类第一视角与机器人数据、用于质量控制和 RL 后训练的密集进度奖励，以及面向真实世界基准的评分量表。
推荐理由
AI 评分 52
Flora 以 125 倍更低的成本达到 Astra 级表现，并给出 500 GPU 日标 20 万小时视频的规模数据，可作为机器人数据引擎的参考。
主题
正文
当前显示原文，中文全文译文尚未就绪。
We built a robotics data engine with Astra-level performance at 125x lower cost. And we are open-sourcing the weights today! Introducing Flora, a suite of open models to parse the physical world and turn raw footage into rich training signals.
Flora already powers our in-house data engine, labeling 200K hours of videos a day on 500 GPUs. We created it to scale the frontier for robot learning. Now you own the weights too.
Flora gives you:
- High-quality, timestamped captions to curate and filter human egocentric and robot data; - Dense progress rewards for quality control and RL post-training; - Rubrics scoring for real-world benchmarking
Try it today, links in thread: