Scale AI and Hugging Face Release FineART Robotics Dataset and Model

Oct 29, 2026
Scale AI and Hugging Face released FineART, an open robotics dataset containing 40,543 recorded episodes and more than 533,000 labeled steps, alongside the FineART-VLA model.

Scale AI and Hugging Face released FineART, an open robotics dataset with subtask labels, alongside the FineART-VLA model, as detailed in a company blog post. The dataset contains 40,543 recorded task episodes covering 1,718 hours of footage across 151 tasks.

FineART includes 533,913 labeled steps, averaging about 13 steps per episode. Each annotation pairs a timed action clip with a natural language description. Pick and place and sorting tasks account for 77 percent of the data, while the remainder covers textiles, cables, tool use, and precision insertion.

FineART-VLA is a vision language action model available through Hugging Face's LeRobot. It uses camera images, language instructions, and robot state data to predict subtasks before converting them into robot actions. Scale AI says the model supports spatial instructions, multistep execution, and transfers across different robot hardware.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Robotics Brief or Daily AI Brief.

Also, consider following us on social media:

AI & Robotics AI Brief AI Brief (X)
Free newsletter

Robotics Brief

Weekly coverage of AI-driven robotics advances in industrial automation, autonomous vehicles, and robotic systems.

Guide
Private humanoid makers by latest disclosed equity round

Private humanoid makers by latest disclosed equity round

Thirteen independent humanoid robot makers, ordered by the size of their latest disclosed round

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.