DataTrain.AI

IntuitivePhysics

WorldBench WorldBench is a new benchmark designed to evaluate the physical understanding and prediction of modern world models…

100K–1M·Images (folder)

Image Multimodal Text

LLaVA-OneVision-1.5-Instruct-Data

LLaVA-OneVision-1.5 Instruction Data Paper Code 📌 Introduction This dataset, LLaVA-OneVision-1.5-Instruct, was collected and integrated during the…

10M–100M·Apache-2.0

Image

Typed Digital Signatures Dataset

10K–100K·MIT

Video

droid

This dataset was created using LeRobot. DROID: A Large-Scale In-the-Wild Robot Manipulation Dataset One of the biggest open-source…

10M–100M·Apache-2.0

Image Multimodal Text

LLaVA-OneVision-2-Data

LLaVA-OneVision-2-Data Training data for the LLaVA-OneVision-2 multimodal model family, covering large-scale video and spatial reasoning corpora used…

<1K·Apache-2.0·Parquet

Multimodal Sensor / Time-series Tabular

T-Rex Dataset

1M–10M·MIT·Parquet

Multimodal Tabular Text

OpenVid-1M

1M–10M·CC-BY·CSV

Text

The Pile

825 GiB of diverse text from 22 curated sources.

>10M·Mixed·Parquet

Audio

LibriSpeech

~1,000 hours of aligned read English speech.

100K–1M·CC-BY·Audio (wav/mp3)

English