datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
GreekMMLU
GreekMMLU
GreekMMLU is a native-sourced benchmark for evaluating massive multitask language understanding in Greek, built from authentic Greek exam-style multiple-choice questions (MCQ) rather than machine-translated English benchmarks.
21,805 questions across 45 subjects
4 high-level groups: STEM, Humanities, Social Sciences, Other
Difficulty/education levels spanning Primary → Secondary → University → Professional (+ an N/A bucket)
Public vs. private split for… See the full description on the dataset page: https://huggingface.co/datasets/dascim/GreekMMLU.audioset-dasheng-0.6b-emb
AudioSet DaSheng-0.6B embeddings
Mean-pooled, float16 embeddings of
danjacobellis/audioset_opus_24kbps
from mispeech/dasheng-0.6B.
Columns
path: source clip path (string)
label: source AudioSet label indices (list of int64)
emb: 1,280-dimensional fixed-size list of float16
Audio is decoded from the source Opus bytes, mixed to mono, and resampled to
16 kHz. The embedding is the model's documented outputdim=None output:
sigmoid applied to the mean of the final… See the full description on the dataset page: https://huggingface.co/datasets/quinnlue/audioset-dasheng-0.6b-emb.quants
QuAnTS: Question Answering on Time Series
QuAnTS is a challenging dataset designed to bridge the gap in question-answering research on time series data.
The dataset features a wide variety of questions and answers concerning human movements, presented as tracked skeleton trajectories.
QuAnTS also includes human reference performance to benchmark the practical usability of models trained on this dataset.
At present, there is no official leaderboard for this dataset.
Dataset… See the full description on the dataset page: https://huggingface.co/datasets/dasyd/quants.dasd-stage1-50k
DASD stage1 - 50k length-filtered subset
A 50,000-example subset of the stage1 (low-temperature) config of
Alibaba-Apsara/Superior-Reasoning-SFT-gpt-oss-120b.
Columns are input / output; output is the verbatim gpt-oss-120b <think> reasoning trace.
How it was built
Started from stage1 (104,829 rows).
Applied the Qwen3-4B-Instruct-2507 chat template and tokenized the full formatted
conversation, then dropped every example over 65,536 tokens (the 64K training… See the full description on the dataset page: https://huggingface.co/datasets/amphora/dasd-stage1-50k.DAS-0910-1This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"observation.imu": {
"dtype": "float32",
"shape": [
6
],
"names": [
"angular_velocity.x",
"angular_velocity.y",
"angular_velocity.z",
"linear_acceleration.x",
"linear_acceleration.y"… See the full description on the dataset page: https://huggingface.co/datasets/fza1796262052/DAS-0910-1.DASS2019_NLP
Dataset Card for DASS2019_NLP
This dataset contains audio and transcript content from DASS2019, the manually transcribed version of the Digital Archive of Southern Speech.
It may be suitable for speech-related NLP processing, modelling, and fine-tuning tasks.
Dataset Details
DASS (Kretzschmar et al. 2012) comprises dialectological interviews with 64 informants conducted between 1968 and 1983; it is a subset of the larger Linguistic Atlas of the Gulf States (LAGS… See the full description on the dataset page: https://huggingface.co/datasets/stcoats/DASS2019_NLP.lelab-test_20260906_162149This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"action": {
"dtype": "float32",
"shape": [
6
],
"names": [
"shoulder_pan.pos",
"shoulder_lift.pos",
"elbow_flex.pos",
"wrist_flex.pos",
"wrist_roll.pos",
"gripper.pos"… See the full description on the dataset page: https://huggingface.co/datasets/dasfas1132/lelab-test_20260906_162149.DASH-B
DASH-B
Object Hallucination Benchmark for Vision Language Models (VLMs)
from the paper
DASH: Detection and Assessment of Systematic Hallucinations of VLMs
Model Evaluation | Citation
Dataset
The benchmark consists of 2682 images for a range of 70 different objects. The used query is "Can you see a object in this image. Please answer only with yes or no." 1341 of the images do not contain the corresponding object but trigger object hallucinations. They were retrieved… See the full description on the dataset page: https://huggingface.co/datasets/YanNeu/DASH-B.llm-perf-dashboardbutterflies_and_moths_vqa
Butterflies and Moths VQA
Dataset Summary
butterflies_and_moths_vqa is a visual question answering (VQA) dataset focused on butterflies and moths. It features tasks such as fine-grained species classification and ecological reasoning. The dataset is designed to benchmark Vision-Language Models (VLMs) for both image-based and text-only training approaches.
Key Features
Fine-Grained Classification (Type1): Questions requiring detailed species identification.… See the full description on the dataset page: https://huggingface.co/datasets/Dasool/butterflies_and_moths_vqa.Apertus-8B-2509-microQAT-logitsThis dataset provides a small sample of TOP-K logits computed using swiss-ai/Apertus-8B-2509 on samples from Data Phase 5 of Apertus pre-training.
Format
This data represents documents packed into chuncks of 4096 tokens separated by EOS. The provided fields are as follows:
input_ids: Input tokens.
index: Positions of top-256 highest-probability next-token predictions for each token.
exp_logits: Normalized probabilities of top-256 highest-probability next-token predictions for each… See the full description on the dataset page: https://huggingface.co/datasets/daslab-testing/Apertus-8B-2509-microQAT-logits.DAS-0910This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"observation.imu": {
"dtype": "float32",
"shape": [
6
],
"names": [
"angular_velocity.x",
"angular_velocity.y",
"angular_velocity.z",
"linear_acceleration.x",
"linear_acceleration.y"… See the full description on the dataset page: https://huggingface.co/datasets/fza1796262052/DAS-0910.DAS-0910-2This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"observation.imu": {
"dtype": "float32",
"shape": [
6
],
"names": [
"angular_velocity.x",
"angular_velocity.y",
"angular_velocity.z",
"linear_acceleration.x",
"linear_acceleration.y"… See the full description on the dataset page: https://huggingface.co/datasets/fza1796262052/DAS-0910-2.bc5cdr-disease-selectionurban_dictionaryKoMultiText
KoMultiText: Korean Multi-task Dataset for Classifying Biased Speech
Dataset Summary
KoMultiText is a comprehensive Korean multi-task text dataset designed for classifying biased and harmful speech in online platforms. The dataset focuses on tasks such as Preference Detection, Profanity Identification, and Bias Classification across multiple domains, enabling state-of-the-art language models to perform multi-task learning for socially responsible AI applications.… See the full description on the dataset page: https://huggingface.co/datasets/Dasool/KoMultiText.trace-dashed-lineThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"action": {
"dtype": "float32",
"names": [
"shoulder_pan.pos",
"shoulder_lift.pos",
"elbow_flex.pos",
"wrist_flex.pos",
"wrist_roll.pos",
"gripper.pos"
],
"shape": [
6… See the full description on the dataset page: https://huggingface.co/datasets/Yunis147/trace-dashed-line.stable-diffusion-promptsSubset dataset of diffusiondb consisting of just unique prompts.
Created this subset dataset for the Prompt Extend project.
mqtt_robot_demo_with_camThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "mqtt_robot",
"total_episodes": 1,
"total_frames": 1282,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:1"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/dashuai2025/mqtt_robot_demo_with_cam.test
Dataset Card for "test"
More Information needed
celebrity-identities
Dataset Card for "celebrity-identities"
More Information needed
DashboardQA
DashboardQA: Benchmarking Multimodal Agents for Question Answering on Interactive Dashboards
🤗Dataset | 🖥️Code | 📄Paper
The abstract of the paper states that:
Dashboards are powerful visualization tools for data-driven decision-making, integrating multiple interactive views that allow users to explore, filter, and navigate data. Unlike static charts, dashboards support rich interactivity, which is essential for uncovering insights in real-world analytical workflows. However… See the full description on the dataset page: https://huggingface.co/datasets/ahmed-masry/DashboardQA.trace-dashed-line_20260820_141731This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"action": {
"dtype": "float32",
"names": [
"shoulder_pan.pos",
"shoulder_lift.pos",
"elbow_flex.pos",
"wrist_flex.pos",
"wrist_roll.pos",
"gripper.pos"
],
"shape": [
6… See the full description on the dataset page: https://huggingface.co/datasets/Yunis147/trace-dashed-line_20260820_141731.indian_cuisine_datasetgymrec__BreakoutNoFrameskip_dash_v4
BreakoutNoFrameskip-v4 Gameplay Dataset
Gameplay recordings (collected by: breakout) from the Gymnasium environment BreakoutNoFrameskip-v4,
captured using gymrec.
Dataset Summary
Stat
Value
Total frames
494,029
Episodes
50
Environment
BreakoutNoFrameskip-v4
Backend
Atari (ALE-py)
Collector(s)
breakout
gymrec version(s)
0.1.0+23e91c8, 0.1.0+72aad18
Environment Configuration
Setting
Value
Frameskip
1
Target FPS
30… See the full description on the dataset page: https://huggingface.co/datasets/tsilva/gymrec__BreakoutNoFrameskip_dash_v4.GUI-Libra-81K-RLDC_inside_comments
DC_inside_comments
This dataset contains 110,000 raw comments collected from DC Inside. It is intended for unsupervised learning or pretraining purposes.
Dataset Summary
Data Type: Unlabeled raw comments
Number of Examples: 110,000
Source: DC Inside
Related Dataset
For labeled data and multi-task annotated examples, please refer to the KoMultiText dataset.
How to Load the Dataset
from datasets import load_dataset
# Load the unlabeled dataset… See the full description on the dataset page: https://huggingface.co/datasets/Dasool/DC_inside_comments.gymrec__BreakoutNoFrameskip_dash_v4_stack4_unroll8_train_readyncbi-disease-selectionhf-coding-tools-dashboard-v2
HuggingFace AI Coding Tools Dashboard (Enhanced)
Enhanced benchmark data from the HuggingFace AI Dashboard — includes query metadata (query_set, intent), run metadata (run_name, run_date), and freshness flags for stale references.
This is the v2 enhanced dataset. The original dataset is at davidkling/hf-coding-tools-dashboard.
Dataset Structure
Split
Description
Rows
results
Enhanced results with query/run metadata and freshness flags
9146
queries… See the full description on the dataset page: https://huggingface.co/datasets/davidkling/hf-coding-tools-dashboard-v2.