Original sourceThe Decoder
Summary
Nvidia has open-sourced Flora, a data engine for annotating robot video data that uses three fine-tuned Qwen3.5 models (9B, 27B and 27B Flora-Rewarder) to segment actions, describe content and assess task success and steps. Nvidia says Flora can annotate up to 200,000 hours of video per day on 512…
Key points
- Robotics and computer vision teams get an open-source video annotation pipeline that lowers the cost of building large datasets.
- Robot training relies heavily on annotated video, and an open-source, low-cost data engine can lower barriers to entry and reshape the data vendor ecosystem.
- Robotics teams can assess annotating video themselves with Flora, but should note that cost comparisons are estimates and real performance depends on hardware and data quality.
Editorial note
This page is Code & Chain's editorial summary of public sources. It may be prepared with AI assistance and published through an automated workflow. Refer to the original sources; this content is not investment, legal, or tax advice.