MONET: A Massive, Open, Non-redundant and Enriched Text-to-image dataset

TL;DR

MONET is a large, open, and richly annotated text-to-image dataset with 104.9 million pairs, designed to facilitate reproducible research in large-scale text-to-image modeling.

Contribution

The paper introduces MONET, a comprehensive, filtered, and augmented dataset for text-to-image tasks, enabling scalable and reproducible research.

Findings

Training on MONET yields competitive GenEval and DPG scores.

MONET's diverse and high-quality data supports large-scale model training.

The dataset accelerates research by providing pre-computed embeddings and annotations.

Abstract

Training large text-to-image models requires high-quality, curated datasets with diverse content and detailed captions. Yet the cost and complexity of collecting, filtering, deduplicating, and re-captioning such corpora at scale hinders open and reproducible research in the field. We introduce MONET, an open Apache 2.0 dataset of approx. 104.9M image--text pairs collected from 2.9B raw pairs across heterogeneous open sources through successive stages of safety filtering, domain-based filtering, exact and near-duplicate removal, and re-captioning with multiple vision-language models covering short to long-form descriptions, and further augmented with synthetically generated samples. Each image is shipped with pre-computed embeddings and annotations to accelerate downstream use. To validate the effectiveness of MONET, we train a 4B-parameter latent diffusion model exclusively on it and reach competitive GenEval and DPG scores, demonstrating that our dataset lowers the barrier to large-scale, reproducible text-to-image research.