Fine-T2I: An Open, Large-Scale, and Diverse Dataset for High-Quality T2I Fine-Tuning

2026-02-10

AI Summary

This work addresses the limitations of existing open-source text-to-image (T2I) fine-tuning datasets—such as low resolution, poor text-image alignment, and insufficient diversity—which significantly hinder the performance of open models compared to proprietary systems. To bridge this gap, the authors construct the first large-scale open-source T2I dataset that approaches pretraining-scale size while maintaining fine-tuning-level quality. The dataset integrates synthetically generated images with real photographs from professional photographers, covering 10 task combinations, 32 prompt categories, 11 visual styles, and 5 prompt templates. Through rigorous multi-stage automated and manual filtering focused on text-image alignment, visual fidelity, and prompt quality, only approximately 5% of the highest-quality samples are retained. Experiments across multiple diffusion and autoregressive models demonstrate substantial improvements in image generation quality and instruction-following capability, effectively narrowing the data divide between open-source and enterprise-grade T2I models.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Generation

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

High-quality and open datasets remain a major bottleneck for text-to-image (T2I) fine-tuning. Despite rapid progress in model architectures and training pipelines, most publicly available fine-tuning datasets suffer from low resolution, poor text-image alignment, or limited diversity, resulting in a clear performance gap between open research models and enterprise-grade models. In this work, we present Fine-T2I, a large-scale, high-quality, and fully open dataset for T2I fine-tuning. Fine-T2I spans 10 task combinations, 32 prompt categories, 11 visual styles, and 5 prompt templates, and combines synthetic images generated by strong modern models with carefully curated real images from professional photographers. All samples are rigorously filtered for text-image alignment, visual fidelity, and prompt quality, with over 95% of initial candidates removed. The final dataset contains over 6 million text-image pairs, around 2 TB on disk, approaching the scale of pretraining datasets while maintaining fine-tuning-level quality. Across a diverse set of pretrained diffusion and autoregressive models, fine-tuning on Fine-T2I consistently improves both generation quality and instruction adherence, as validated by human evaluation, visual comparison, and automatic metrics. We release Fine-T2I under an open license to help close the data gap in T2I fine-tuning in the open community.

Problem

• text-to-image
• fine-tuning
• dataset
• data gap
• open research

Innovation

• text-to-image
• fine-tuning dataset
• high-quality data
• open dataset
• instruction adherence