Skip to content

Dataset · daimonrobotics

Omni-Modal Dataset for Physical AI

daimonrobotics/Daimon-Infinity

Over 10,000 hours of VTLA data with tactile sensors for robot training.

What's inside

Daimon-Infinity is positioned as the largest omni-modal robotics dataset for physical AI. The card indicates that the project opens up over 10,000 hours of vision-tactile-language-action (VTLA) data, including tactile sensing. Languages are English and Chinese; the goal is to build a tactile-centric data ecosystem. The format is not specified in the card.

What you can do with it

  • Train a robot VLA model on vision-tactile-language-action data
  • Fine-tune LoRA for fine manipulation policies with tactile sensors
  • Evaluate an omni-modal model on physical AI tasks
Size
1K–10K rows
Downloads this month
114.1K
Likes
113

How to load it

from datasets import load_dataset

ds = load_dataset("daimonrobotics/Daimon-Infinity")