Dataset · LucasFang
6M images for text-to-image reasoning
LucasFang/FLUX-Reason-6M
6 million images generated by FLUX.1-dev and 20 million English-Chinese descriptions for training generative models.
What's inside
A large-scale text-to-image dataset created to enhance complex reasoning in generative models. It contains 6 million images synthesized by FLUX.1-dev and 20 million bilingual descriptions in English and Chinese. The data focuses on high quality and reasoning tasks, with image and text modalities.
What you can do with it
- Fine-tune or adapt a text-to-image model with reasoning
- Train LoRA for complex prompts in English and Chinese
- Evaluate generative models on reasoning-based descriptions
- Size
- 1M–10M rows
- Downloads this month
- 15.3K
- Likes
- 122
Models trained on it
- caikybaldo999/zyi-256-flashText → image 602
- Logolabs/agate-preview-003Text → image 484
- caikybaldo999/zyi-flashText → image 196
- ivanmikhnenkov/tinydit-256Text → image 152
- SupraLabs/Supra2-IMGText → image 116
- Logolabs/agate-preview-001Text → image 105
- cpral/HomeIMG-1.2B-PL-EN-epoch49.2-DanceOPD2048Text → image 92
- cpral/HomeIMG-1.2B-PL-EN-epoch49.2Text → image 76
How to load it
from datasets import load_dataset
ds = load_dataset("LucasFang/FLUX-Reason-6M")