datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
BlenderLore
The target scope is 22,360 video-associated Blender project instances, not 22,360 distinct tutorial videos. Uploads are in progress, so the currently published files may be a subset of this target. The 44 biomedical project instances and one software-bundled Dome template are excluded.
Data Structure
The dataset is organized as a collection of sample-level directories under assets/. Each directory corresponds to one Blender creation task and follows the structure below:… See the full description on the dataset page: https://huggingface.co/datasets/FreedomIntelligence/BlenderLore.cad-gen-freecad
CAD Generation Dataset
Each row in this dataset describes one parametric CAD part. Columns:
id — row identifier (also the basename of the per-row asset files)
name — part family (e.g. flanges, spur_gear_stock)
description — natural-language description of the geometry
key_parameters — the dimensions that drive the parametric model
image — 512×512 PNG preview rendered from the FCStd
fcstd_path — relative path inside this repo to the parametric FreeCAD document (fcstd/<id>.FCStd)… See the full description on the dataset page: https://huggingface.co/datasets/gnucleus-ai/cad-gen-freecad.PubMedVision
News
[2025/02/18]: We add the original captions of PubMedVision in PubMedVision_Original_Caption.json, as well as the Chinese version of PubMedVision in PubMedVision_Chinese.json.
[2024/07/01]: We add annotations for 'body_part' and 'modality' of images, utilizing the HuatuoGPT-Vision-7B model.
PubMedVision
PubMedVision is a large-scale medical VQA dataset. We extracted high-quality image-text pairs from PubMed and used GPT-4V to reformat them to enhance their quality.… See the full description on the dataset page: https://huggingface.co/datasets/FreedomIntelligence/PubMedVision.SciFigPlag-Bench
SciFigPlag-Bench: A Benchmark for Provenance-Aware Scientific Figure Plagiarism Detection
🌐 Homepage |
📖 arXiv
SciFigPlag-Bench is a benchmark for provenance-aware scientific figure plagiarism detection. It evaluates whether a suspicious figure reuses evidence from a specific source figure, which figure is the original source, how the reused content has been transformed, and where the reused evidence appears.
The benchmark is designed to evaluate vision-language models… See the full description on the dataset page: https://huggingface.co/datasets/FreeLand123/SciFigPlag-Bench.ALLaVA-4V
📚 ALLaVA-4V Data
Generation Pipeline
LAION
We leverage the superb GPT-4V to generate captions and complex reasoning QA pairs. Prompt is here.
Vison-FLAN
We leverage the superb GPT-4V to generate captions and detailed answer for the original instructions. Prompt is here.
Wizard
We regenerate the answer of Wizard_evol_instruct with GPT-4-Turbo.
Dataset Cards
All datasets can be found here.
The structure of naming is shown below:
ALLaVA-4V
├──… See the full description on the dataset page: https://huggingface.co/datasets/FreedomIntelligence/ALLaVA-4V.FreeStyle_Dataset
0426 CRef/SRef LoRA Triplet Dataset
This dataset contains CRef/SRef LoRA triplets exported from the 0426 diffusion training data. Each training example has three images:
content: content reference image, used as cref_0
style: style reference image, used as sref_0
target: image generated from the combined content + style condition
Use triplets.csv as the main entry point. Image-level CSV files are provided only for deduplicated metadata and provenance lookup.… See the full description on the dataset page: https://huggingface.co/datasets/Blue2Giant/FreeStyle_Dataset.freevocab-scannet-scannetpp-outputfreeeternalsummer
Bangumi Image Base of Free! -eternal Summer-
This is the image base of bangumi Free! -Eternal Summer-, we detected 24 characters, 2471 images in total. The full dataset is here.
Please note that these image bases are not guaranteed to be 100% cleaned, they may be noisy actual. If you intend to manually train models using this dataset, we recommend performing necessary preprocessing on the downloaded dataset to eliminate potential noisy samples (approximately 1% probability).… See the full description on the dataset page: https://huggingface.co/datasets/BangumiBase/freeeternalsummer.omnidocbench-render-compare
OmniDocBench Render-and-Compare
This dataset contains the rendered HTML reconstructions and comparison images produced
by a render-and-compare pipeline — a reference-free visual similarity evaluation
framework for OCR systems.
Overview
The pipeline processes each page of OmniDocBench through
a Qwen3.5-122B-A10B OCR model, renders the structured output back to a PNG via HTML
(reconstructed.png), and compares it against the original page scan (masked_original.png)
using… See the full description on the dataset page: https://huggingface.co/datasets/gt-free-ocr-metrics/omnidocbench-render-compare.FreeArt-21
Paper
FreeArtGS: Articulated Gaussian Splatting Under Free-moving Scenariohttps://huggingface.co/papers/2603.22102
license: apache-2.0
FreeGave-GoPro
FreeGave-GoPro Dataset
This dataset is proposed by FreeGave.
Structure
The structure of the dataset is as:
DynObjects
| - data
| | - pen1:
| | | - train: serves as training data
| | | - val: used for evaluating novel view interpolation
| | | - test: used for evaluating future extrapolation
| | | - transforms_train.json: camera poses and other meta informations for training set
| | | - transforms_val.json: camera poses and other meta informations for novel view… See the full description on the dataset page: https://huggingface.co/datasets/scintigimcki/FreeGave-GoPro.FreeSplatterStaticMedical_Multimodal_Evaluation_Data
Evaluation Guide
This dataset is used to evaluate medical multimodal LLMs, as used in HuatuoGPT-Vision. It includes benchmarks such as VQA-RAD, SLAKE, PathVQA, PMC-VQA, OmniMedVQA, and MMMU-Medical-Tracks.
To get started:
Download the dataset and extract the images.zip file.
Find evaluation code on our GitHub: HuatuoGPT-Vision.
This open-source release aims to simplify the evaluation of medical multimodal capabilities in large models. Please cite the relevant benchmark… See the full description on the dataset page: https://huggingface.co/datasets/FreedomIntelligence/Medical_Multimodal_Evaluation_Data.FreeStyle_Bench
FreeStyle Bench
This repository contains benchmark inputs for evaluating FreeStyle reference-based image editing.
It includes two benchmark subsets:
sref_bench/: a style-transfer benchmark.
cref_sref_bench/: a content-and-style dual-reference benchmark.
Both subsets use the same basic layout: a content-reference folder, a style-reference folder, and a prompts.json file that links each pair to its text prompt.
Repository Layout
<repo-root>/
README.md… See the full description on the dataset page: https://huggingface.co/datasets/Blue2Giant/FreeStyle_Bench.FreeGraspData Free-form language-based robotic reasoning and grasping Dataset: FreeGraspData
Homepage: https://tev-fbk.github.io/FreeGrasp/
Paper: https://arxiv.org/abs/2503.13082
FreeGrasp Code: https://github.com/tev-fbk/FreeGrasp_code
NOTE: the scene ids and objects ids are given from MetaGraspNet v2
The objects instance segmentation maps are available at https://huggingface.co/datasets/FBK-TeV/FreeGraspData/blob/main/data/npz_file.zip
Dataset Description
Examples of FreeGraspData at… See the full description on the dataset page: https://huggingface.co/datasets/FBK-TeV/FreeGraspData.FreeArt3D
Dataset Card for Dataset Name
Preprocessed dataset of PartNet-Mobility objects for FreeArt3D.
Dataset Sources
Repository:: https://github.com/CzzzzH/FreeArt3D
Paper: https://huggingface.co/papers/2510.25765
Demo: https://huggingface.co/spaces/MorPhLingXD/FreeArt3D
BibTeX:
@InProceedings{chen2025freeart3d,
title = {FreeArt3D: Training-Free Articulated Object Generation using 3D Diffusion},
author = {Chen, Chuhao and Liu, Isabella and Wei, Xinyue and Su, Hao and… See the full description on the dataset page: https://huggingface.co/datasets/MorPhLingXD/FreeArt3D.kodakThe pictures below link to lossless, true color (24 bits per pixel, aka "full
color") images. It is my understanding they have been released by the Eastman
Kodak Company for unrestricted usage. Many sites use them as a standard test
suite for compression testing, etc. Prior to this site, they were only
available in the Sun Raster format via ftp. This meant that the images could
not be previewed before downloading. Since their release, however, the lossless
PNG format has been incorporated into all the major browsers. Since PNG
supports 24-bit lossless color (which GIF and JPEG do not), it is now possible
to offer this browser-friendly access to the images.TCM-Vision-Benchmark
📚 Introduction
This is the text benchmark for ShizhenGPT, a multimodal LLM for Traditional Chinese Medicine (TCM).
For details, see our paper and GitHub repository.
📊 Benchmark Overview
The benchmark is composed of 7 sections, each compiled from different authoritative TCM illustrated books.
Samples
TCM Patent
1119
TCM Material
1020
TCM Herb
1100
Tongue768
Palm
640
Holism
1011
Tuina
831
Eye
715
⚒️ Data Construction
{… See the full description on the dataset page: https://huggingface.co/datasets/FreedomIntelligence/TCM-Vision-Benchmark.FreeCHROMAk12-freeformcaptcha
best.pth
model = CaptchaCNNTransformer(
img_h=32,
img_w=128,
dim=256,
depth=6,
heads=4,
num_classes=num_classes,
channels=1,
dropout=0.2,
).to(device)
free-to-use-pixelart
Free-to-use Pixel Art
Dataset Details
This dataset was collected on 25th May, 2024.
It's a small subset of the free-to-use images on PixilArt.
At the time of publication, this dataset was covered by permissive terms that allow commercial use.
Dataset Description
This dataset is unique in that it contains the pixel group size for each collected sample, which might assist in experiments on microconditioning inputs on an adapter to control this value of the unit… See the full description on the dataset page: https://huggingface.co/datasets/bghira/free-to-use-pixelart.trocr-hebrew-freefonts-BYFreeFixFreegraspData_V2This includes the source data of UnoGrasp, they are selected from MetaGraspNetV2 from 37 vpt to 4 vpt.
It has no practical function and is only used for tracing the source.
heb-freefonts-250kfreeFreeCAD_Sketches_Pics
🧩 FreeCAD Sketch Python Dataset
This dataset contains approximately 1,000 Python files and their corresponding images defining parametric FreeCAD sketches. Each python file represents a geometric sketch scripted using the FreeCAD Python API.
The dataset is intended for training and fine-tuning large language models (LLMs) and vision large language models (vLLMs) and other AI systems to understand and generate CAD-based parametric geometry code, particularly for text-to-CAD and… See the full description on the dataset page: https://huggingface.co/datasets/Yas1n/FreeCAD_Sketches_Pics.bangladesh-urban-traffic-free-sample-v2
Bangladesh Urban Traffic Dataset (Dhaka Mixed Traffic Dataset) — Free Sample
This Bangladesh Traffic Dataset contains real-world mixed traffic scenes including rickshaws, motorcycles, buses, cars, pedestrians, market congestion, intersections, and complex urban mobility patterns.
The dataset is designed for computer vision, object detection, autonomous driving, ADAS, robotics, perception systems, physical AI, traffic analytics, and emerging-market mobility research.
This… See the full description on the dataset page: https://huggingface.co/datasets/origindatalab/bangladesh-urban-traffic-free-sample-v2.trocr-hebrew-freefonts9