datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
flat-pack-bench
Flat-Pack Bench 🧩
Furniture assembly as a spatio-temporal stress test for large vision-language models.
Flat-Pack Bench is a multiple-choice benchmark for evaluating fine-grained
spatio-temporal understanding in real furniture assembly videos. Each question
asks a model to reason about object parts, contact events, assembly order, final
connectivity, or part identity across time.
Project page: https://flat-pack-bench.github.io
🎯 Benchmark Tasks
The benchmark… See the full description on the dataset page: https://huggingface.co/datasets/justachetan/flat-pack-bench.vqa-rad
Dataset Card for VQA-RAD
Dataset Description
VQA-RAD is a dataset of question-answer pairs on radiology images. The dataset is intended to be used for training and testing
Medical Visual Question Answering (VQA) systems. The dataset includes both open-ended questions and binary "yes/no" questions.
The dataset is built from MedPix, which is a free open-access online database of medical images.
The question-answer pairs were manually generated by a team of clinicians.… See the full description on the dataset page: https://huggingface.co/datasets/flaviagiammarino/vqa-rad.path-vqa
Dataset Card for PathVQA
Dataset Description
PathVQA is a dataset of question-answer pairs on pathology images. The dataset is intended to be used for training and testing
Medical Visual Question Answering (VQA) systems. The dataset includes both open-ended questions and binary "yes/no" questions.
The dataset is built from two publicly-available pathology textbooks: "Textbook of Pathology" and "Basic Pathology", and a
publicly-available digital library: "Pathology… See the full description on the dataset page: https://huggingface.co/datasets/flaviagiammarino/path-vqa.vision-flan_191-task_1k
🚀 Vision-Flan Dataset
vision-flan_191-task-1k is a human-labeled visual instruction tuning dataset consisting of 191 diverse tasks and 1,000 examples for each task.
It is constructed for visual instruction tuning and for building large-scale vision-language models.
Paper or blog for more information:
https://github.com/VT-NLP/MultiInstruct/
https://vision-flan.github.io/
Paper coming soon 😊
Citation
Paper coming soon 😊. If you use Vision-Flan, please use the… See the full description on the dataset page: https://huggingface.co/datasets/Vision-Flan/vision-flan_191-task_1k.EmbSpatial-Bench
Introduction
Disclaimer: This dataset is organized and adapted from Phineas476/EmbSpatial-Bench. The original data was image format and has been converted here into a more accessible and easy-to-use format.
EmbSpatial-Bench is a benchmark for evaluating embodied spatial understanding of LVLMs. The benchmark is automatically derived from embodied scenes and covers 6 spatial relationships from an egocentric perspective. The constructed benchmark comprises a total of 3,640 QA pairs… See the full description on the dataset page: https://huggingface.co/datasets/FlagEval/EmbSpatial-Bench.ERQA
Introduction
Disclaimer: This dataset is organized and adapted from embodiedreasoning/ERQA. The original data was provided in TFRecord format and has been converted here into a more accessible and easy-to-use format.
This evaluation benchmark covers a variety of topics related to spatial reasoning and world knowledge focused on real-world scenarios, particularly in the context of robotics. Please find more details and visualizations in the tech report.
Data Fields… See the full description on the dataset page: https://huggingface.co/datasets/FlagEval/ERQA.nih-chest-xray-14-flatosworld-glm-5.3-flash-trajThese are the trajectory results from our GLM-5.3-Flash evaluation on OSWorld.
For detailed evaluation results, configuration, and additional information, please refer to the following GitHub issue:
https://github.com/xlang-ai/OSWorld/issues/591
word-flag-dataWhere2Place
Introduction
Disclaimer: This dataset is organized and adapted from wentaoyuan/RoboPoint. The original data was image format and has been converted here into a more accessible and easy-to-use format.
This dataset contains 100 real-world images to evaluate free space reference using spatial relations. The images are collected from various cluttered environments. Each image is labeled with a sentence describing the desired some free space and a mask of the desired region.… See the full description on the dataset page: https://huggingface.co/datasets/FlagEval/Where2Place.NSD-Flat
NSD-Flat
[GitHub] [🤗 Hugging Face Hub]
A Hugging Face dataset of pre-processed brain activity flat maps from the Natural Scenes Dataset, constrained to a visual cortex region of interest and rendered as PNG images.
Load the dataset
Load the dataset from Hugging Face Hub
from datasets import load_dataset
dataset = load_dataset("clane9/NSD-Flat", split="train")
Building the dataset
1. Download source data
Run download_data.sh to download the… See the full description on the dataset page: https://huggingface.co/datasets/clane9/NSD-Flat.MICCAI_FLARE_diabetic_retinopathyFlareBench
FlareBench Dataset
This dataset contains solar observation data for solar flare prediction research, specifically focusing on the year 2011.
Dataset Structure
solar_images/
└── aia/
├── 2011/
├── 2012/
├── 2013/
└── ...
Data Description
solar_images/aia/: AIA (Atmospheric Imaging Assembly) solar images from SDO
Multi-wavelength solar observations
Organized by year and date
FITS format files with calibrated data
License
This… See the full description on the dataset page: https://huggingface.co/datasets/sh237/FlareBench.sam3d-flat-20260329-022951savtadepth-flags-V2twinicl-bench
TwinICL
38 tasks, each with 132 underlying examples rendered in eight variants: 40,128 rows in total.
Each row contains only:
task: a readable task name.
variant: the text style or image palette.
input_text: the text input, or null for image examples.
input_image: the image input, or null for text examples.
answer: the expected text answer.
The eight variants are lowercase/comma, lowercase/semicolon, uppercase/comma, uppercase/semicolon, and images in neutral, warm, cool, and… See the full description on the dataset page: https://huggingface.co/datasets/lab-flair/twinicl-bench.vision-flan
Image generated by https://ideogram.ai/
We introduce Vision-Flan, the largest human-annotated visual instruction tuning dataset that consists of 200+ diverse vision-language tasks derived from 101 open-source computer vision datasets. Each task is equipped with an expert written instruction and carefully designed templates for the inputs and outputs. The dataset encompasses a wide range of tasks such as image captioning, visual question-answering, and visual understanding. Vision-Flan is… See the full description on the dataset page: https://huggingface.co/datasets/Vision-Flan/vision-flan.conceptual-12m-mbart-50-multilingualflat-pack-bench-misc
🛠️ Flat-Pack Bench Misc
This repository contains auxiliary artifacts for Flat-Pack Bench: ablation questions with scrambled part IDs, corresponding prompt-mask variants, cached TVA segmentation tracks, and TVA agent traces. The main benchmark data and evaluation outputs live in the companion Flat-Pack Bench repositories; this repo keeps the heavier or analysis-specific assets separate.
Repository Map
Path
Contents
scrambled-questions/
Base and seeded… See the full description on the dataset page: https://huggingface.co/datasets/justachetan/flat-pack-bench-misc.MeasureBench
Do Vision-Language Models Measure Up? Benchmarking Visual Measurement Reading with MeasureBench
🏠Project Page | 💻Code | 📖Paper | 🤗Data
Fine-grained visual understanding tasks such as visual measurement reading have been surprisingly challenging for frontier general-purpose vision-language models. We introduce MeasureBench, a benchmark with diverse images of measuring instruments collected from both real-world images and a new data synthesis pipeline.
MeasureBench comprises 2442… See the full description on the dataset page: https://huggingface.co/datasets/FlagEval/MeasureBench.purple_flarerico_sca_refexp_synthetic_flat
Dataset Card for "rico_sca_refexp_synthetic_flat"
More Information needed
flair
Federated Learning Annotated Image Repository (FLAIR): A large labelled image dataset for benchmarking in federated learning
FLAIR was published at NeurIPS 2022 (paper)
(Preferred) Benchmarking FLAIR is available in pfl-research (repo, paper).
The ml-flair repo contains a setup for benchmarking with TensorFlow Federated and notebooks for exploring data.
FLAIR is a large dataset of images that captures a number of characteristics encountered in federated learning (FL) and… See the full description on the dataset page: https://huggingface.co/datasets/apple/flair.ROBIN-ImagesGT-Merged-Flanora-AI-v1
Flanora AI/ROBIN-ImagesGT-Merged-Flanora-AI-v1
ROBIN-ImagesGT-Merged-Flanora-AI-v1 is a curated collection of 622 floor-plan images created by merging floor-plan data from the ROBIN dataset and the CVC-FP / ImagesGT dataset.
The dataset is organized into six bedroom-count categories: 0 bedroom, 1 bedroom, 2 bedroom, 3 bedroom, 4 bedroom, and 5 bedroom.
The dataset contains the original, unprocessed floor-plan images. No image preprocessing or transformation was applied to the… See the full description on the dataset page: https://huggingface.co/datasets/BJyotibrat/ROBIN-ImagesGT-Merged-Flanora-AI-v1.VideoChat-Flash-Training-Data-subsetregs-flat2danimerge-v20conceptual-captions-12This file contains English captions from Conceptual 12M dataset by Google. Since we don't own the images, we have provided the link to images, name of downloaded file, and caption for that image in the TSV file.
We would like to thank Luke Melas for helping us get the cleaned CC-12M data on our TPU-VMs.
FireDetectionDataset-flame-forest-flameye-wildfire
FlamEye — Wildfire Detection Dataset
A merged, deduplicated, and augmented dataset for real-time wildfire detection (fire and smoke) from CCTV/surveillance cameras. Built to train YOLOv8m for early-stage fire detection.
Classes
ID
Name
0
fire
1
smoke
Dataset Statistics
Split
Images
Train
~10,929
Validation
~3,000
Test
~1,500
Sources
Dataset
Source
Notes
D-Fire
Kaggle
Class IDs remapped:… See the full description on the dataset page: https://huggingface.co/datasets/baizhanquan/FireDetectionDataset-flame-forest-flameye-wildfire.GaME_Flatrealsense-multicam-tabletop
RealSense Multi-Camera Tabletop
Synchronized multi-view RGB + stereo IR captures of a tabletop scene from 4 Intel RealSense
cameras, recorded with fixed camera positions. Includes FoundationStereo depth for two scenes
and chessboard-derived extrinsics for merging the views into a single point cloud.
Layout
scene_000NN/
├── camera_poses.json # extrinsics, only in calibration scenes (see table)
└── <camera_serial>/
├── rgb/00000.jpg ... 00119.jpg #… See the full description on the dataset page: https://huggingface.co/datasets/Flashkernel/realsense-multicam-tabletop.