data pipelines

Nvidia's open-source Flora tool labels an hour of robot video for about 72 cents, compared with roughly $88.50 using GPT-6 Astra, a dramatic cost difference for robotics training data pipelines. The claim highlights growing pressure on expensive proprietary models as open-source alternatives deliver comparable labeling work at a fraction of the price, a gap robotics developers will likely weigh when scaling data annotation.

Nvidia has open-sourced Flora, a tool that automatically labels robot training videos. According to reports, it cuts labeling costs to roughly 125 times less than using OpenAI's GPT-6 Astra. The move gives robotics developers a low-cost way to prepare video data for training models, and it puts Nvidia in direct competition with proprietary AI services on a key step of the robot learning pipeline.