3D scenes for robotics in Isaac Sim
Reconstructed 3D scenes in USD/USDZ for training and testing end-to-end robots in NVIDIA Isaac Sim.
Hugging Face datasets with plain-language descriptions: what's inside and what models to train on them. 4,788 in the catalog so far, with new ones added every hour.
Answered by Gemini 4 · the chat is kept only in this tab
Reconstructed 3D scenes in USD/USDZ for training and testing end-to-end robots in NVIDIA Isaac Sim.
Benchmark of 700+ real-world images with questions and verifiable answers for multimodal models.
306,207 3D brain MRI scans with clinical data for self-supervised learning.
The first benchmark for evaluating recognition and parsing of real digital and photographed documents across 6 languages.
Around 13 million PNG images across ~19,000 classes with new author-generated captions.
1,148 graphic layouts with detailed markup of layers, typography, vectors, and styles.
108,501 facial images annotated by race, gender, and age to combat bias in computer vision.
Laboratory test results for proteins generated by the Claude model to bind scorpion toxins.
Large dataset of 3D brain MRIs for unsupervised pretraining of medical models.
Web-optimized 3D organ and body system assemblies in GLB format with labeled structures.
Question-answer pairs for MedPix medical images for VQA training and testing.
708k English news articles from 2017–2019 sourced from Common Crawl.
The first major open dataset for Ukrainian handwritten text recognition spanning over a century.
A large synthetic dataset of Japanese personas covering demographics, geography, and personality traits.
Simulator scenes, model responses, and step-by-step evaluations to test the reliability of multimodal models as judges.
A set of 31,500 RGB images for classifying Earth remote sensing scenes.
A reference image dataset annotated with WordNet concept hierarchies for training and testing image classification models.
About 4,000 verified examples of generating Three.js 3D models with reasoning and renders.
Dataset with military aviation images and bounding box annotations for 43 aircraft types.
Over 20 thousand safe-rated anime illustrations with detailed tags, emotions, and descriptions in three languages.
Over half a million vector graphic comparisons of 75 models across text prompts with 3.4 million human votes.
A LAION/CC/SBU sub-sample with concept balance and synthetic BLIP captions for LLaVA pre-training.
127K query-image pairs for training visual document retrieval.
CLIP-filtered dataset of 100k to 1M examples for training text-based photo editing.