00:00:00

The future needs data from everyday people.

Training-ready multimodal human-activity datasets for physical AI. Built by international and national gold medalists in math and robotics.

Request free sample dataset
World's Most Diverse
Sample Dataset
50Hours of data
60+Tasks
20+Environments

Every~2–5 min

is unique in either task or environment

Every physical AI company needs a foundation model first.

Before any task-specific training, robots learn broad manipulation, navigation, hand-object interaction and spatial understanding. That's 90–99% of their training data. It has to come from real people.

Founded by NTU robotics researchers.

RealAI was started by researchers from Nanyang Technological University in Singapore, and works with NTU's MARS Lab on physical AI.

#1NTU in AI worldwide
ShanghaiRanking
Top 5NTU in Data Science & AI
QS World Rankings
Research partner

NTU MARS Lab

NVIDIA-backed research in physical AI, multimodal LLMs and robotics.

  • NVIDIA Academic Grant for multimodal robot manipulation, 2026
  • Papers at ICLR, CVPR, ICRA and RSS
  • ICLR oral, top 1.8%
  • Meta Project Aria participant
  • ECCV challenge winner

MARS Lab collaborators

  • Sony
  • NVIDIA
  • Sharpa
  • MIT
  • ETH Zurich
  • Collaborator
  • Collaborator
  • HarvardHarvard
  • UC BerkeleyUC Berkeley
  • StanfordStanford
  • OxfordOxford
  • PrincetonPrinceton

No saturated tasks.

For the first time ever, capture isn't happening in staged rooms and warehouses, with the same tasks repeated until the hours pile up. Models can now learn in homes people actually live in.

Typical capture dataset

A few tasks hold most of the hours.

Most recorded taskLeast
RealAI

Hours stay balanced across every task.

Most recorded taskLeast

Multimodal data capture.

We record video, hand motion and 3D room data at the same time, from the wearer's point of view. All streams are synced.

Request sample data
Visual
01 / Visual

Visual

The task from eye level, exactly as it happened.

Egocentric RGB
02 / + Motion

Motion

Every finger joint, grip and reach, tracked through the whole task. The fine dexterity robots need to learn.

Hand pose · body pose
03 / + Spatial

Spatial

Room geometry and object positions, mapped as the person moves.

Depth · SLAM · 3D scene

We own the entire pipeline.

The hardware, the app and the data engine are all built and run in-house.

  1. RealAI wearable capture device with stereo cameras01 / Wear
    Custom hardware

    Built in-house.

    Contributors wear our hardware and go about their day at home.

  2. RealAI contributor app showing today's recordings02 / Record
    RealAI app

    Record everyday tasks.

    The app assigns tasks, so every gap in the dataset gets filled on purpose.

  3. RawProcessed03 / Process
    Data engine

    Turn it into data.

    Every session is synced, privacy-blurred and labeled.

  4. 04 / Train
    Foundation models

    Train physical AI.

    Models learn from real people in real homes.

Quality at scale.

Our hardware is built for mass production, so hours grow with demand and quality holds.

88%

Usable data rate.

UsableExcluded
4 months

From idea to production and a sample dataset.

Build physical AI on real-world data.

data@realai.___