datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
p2-etf-amortised-variational-rl-resultslunara-aesthetic-image-variations
Dataset Card for Moonworks Lunara Aesthetic II
This dataset introduces the second open-source release by Moonworks. This dataset contains original image and art created by Moonworks and their contextual variations generated by Moonworks Lunara, a sub-10B parameter model with a novel diffusion mixture architecture.
Paper: https://arxiv.org/pdf/2602.01666
While part 1 is intended for learning and evaluating regional as well as region-agnostic art styles, part 2 is intended for… See the full description on the dataset page: https://huggingface.co/datasets/moonworks/lunara-aesthetic-image-variations.wikipedia-trivia-query-variationimagenet-variations-synth-pilot
ImageNet Variations Synth Pilot (1K)
Synthetic vision-language instruction pilot from laion/imagenet_variations prompts.
Pipeline
Sample 1000 prompts from imagenet_variations
Add one style framing per prompt (website / textbook / diagram / …)
Generate image with FLUX.1-schnell
Encode with SEED-2 (<seed2_N>, 32 tokens)
Qwen2.5-VL-7B-Instruct looks at the actual image and writes instruction + response(modes: describe / QA / howto / reverse)
Format:… See the full description on the dataset page: https://huggingface.co/datasets/EmpathicRobotics/imagenet-variations-synth-pilot.rlbench_peract_variation0view_variation_finalvariations-merge-test-4
Merged LeRobot Dataset
This dataset was created by merging multiple LeRobot datasets using the LeRobot merge tool.
Source Datasets
This merged dataset combines the following 2 datasets:
jackvial/koch_screwdriver_attach_orange_panel_rse14clean
jackvial/koch_screwdriver_attach_orange_panel_e125
Dataset Statistics
Total Episodes: 139
Total Frames: 26038
Robot Type: koch_screwdriver_follower
FPS: 30
Dataset Structure
meta/info.json:
{… See the full description on the dataset page: https://huggingface.co/datasets/jackvial/variations-merge-test-4.Libra_bimanual_fairino_plug_in_container_r_a_angle_variationsThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "FAIRINO_BIMANUAL_SIM",
"total_episodes": 95,
"total_frames": 20345,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 15,
"splits": {
"train": "0:95"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/Libra-TeV/Libra_bimanual_fairino_plug_in_container_r_a_angle_variations.ensembl_variations
Ensembl Variations (Parquet Format)
This dataset contains Ensembl human genetic variations converted to Parquet format for fast and efficient VCF annotation.
Usage
With Polars (Recommended)
import polars as pl
# Load variants for chromosome 21
df = pl.scan_parquet("hf://datasets/just-dna-seq/ensembl_variations/data/homo_sapiens-chr21.parquet")
# Filter variants by position
variants = df.filter(
(pl.col("POS") >= 10000000) & (pl.col("POS") <= 20000000)… See the full description on the dataset page: https://huggingface.co/datasets/just-dna-seq/ensembl_variations.libero_spatial_visual_variations
LIBERO Spatial Visual Variations
This derived dataset expands libero_spatial with six visual conditions per
source trajectory: the original LIBERO appearance plus five scene-level visual
palettes. It preserves the original task states, actions, rewards, dones,
robot state vectors, proprioceptive observations, task language, and BDDL task
semantics.
Visual conditions:
original: Original - Default LIBERO spatial visual appearance.
nordic-sage: Nordic Sage - White marble, pale plaster… See the full description on the dataset page: https://huggingface.co/datasets/ccwatson/libero_spatial_visual_variations.wikipedia-nq-query-variationimagenet-variations-synth-pilot-v2
ImageNet Variations Synth Pilot v2 (1K)
Second pilot after Huu feedback: expanded instruction modes + noisier-prompt filtering.
Pipeline
Flux.1-schnell → Seed2 tokens → Qwen2.5-VL-7B on the actual image → USER/ASSISTANT.
Modes (weighted)
fine_qa, describe, qa, mood, people, generate, howto, reverse (discriminative, low weight).
Format
USER: <seed2> ... </seed2> {instruction} ASSISTANT: {response}
Contents… See the full description on the dataset page: https://huggingface.co/datasets/EmpathicRobotics/imagenet-variations-synth-pilot-v2.kokoro-hindi-voice-variations
Kokoro Hindi Voice Variations
85 PCA-based voice variations of hf_alpha (Hindi female) for Kokoro-82M TTS.
How these were made
PCA on all 54 official Kokoro voicepacks
Projected hf_alpha into PCA space
Shifted individual PCs and combinations
Hindi phonetics preserved (all start from hf_alpha)
Files
*.npy — Voicepack tensors [510, 1, 256] float32
*.ogg — Sample audio (Hindi text)
catalog.json — Metadata
Single PC shifts (60): pc{1-10}_{±5,±15,±25}… See the full description on the dataset page: https://huggingface.co/datasets/hf4uwho/kokoro-hindi-voice-variations.frazier_dataset_v2_object_variation_20260916_143913This 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/seriintan/frazier_dataset_v2_object_variation_20260916_143913.frazier_dataset_v2_object_variationThis 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/seriintan/frazier_dataset_v2_object_variation.eval_rlintern-wettissue-pick_variation-200_smolvlaThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "so_follower",
"total_episodes": 22,
"total_frames": 32026,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:22"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/yeeunleee/eval_rlintern-wettissue-pick_variation-200_smolvla.variations-merge-test
Merged LeRobot Dataset
This dataset was created by merging multiple LeRobot datasets using the LeRobot merge tool.
Source Datasets
This merged dataset combines the following 2 datasets:
jackvial/koch_screwdriver_attach_orange_panel_rse14clean
jackvial/koch_screwdriver_attach_orange_panel_lse23clean
Dataset Statistics
Total Episodes: 37
Total Frames: 7233
Robot Type: koch_screwdriver_follower
FPS: 30
Dataset Structure
meta/info.json:
{… See the full description on the dataset page: https://huggingface.co/datasets/jackvial/variations-merge-test.artificial_variationsetsThis dataset contains artificial variation sets generated by GPT-4o-mini.Variation sets are sets of (mostly consecutive) utterances that convey a similar intent with slight variations in word choice and structure (Küntay and Slobin, 1996). They are a characteristic feature of Child-Directed Speech.
All artificial variation sets are available in artificial-variationsets.txt .
Described in the following paper: https://arxiv.org/abs/2411.09587
In this paper, artificial variation sets were… See the full description on the dataset page: https://huggingface.co/datasets/akari000/artificial_variationsets.lerobot_rlbench_rvt_push_buttons_variation0rollout_turbovla_object_variation_close_curtain_20260919_105323This 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/seriintan/rollout_turbovla_object_variation_close_curtain_20260919_105323.imagenet-variations-synth-pilot-v3
ImageNet Variations Synth Pilot v3.1 (diverse, 1K)
Huu + Claude multimodal instruction pipeline — quality-fixed re-run.
Pipeline
Flux.1-schnell (Huu style phrasings + visual style boosts + aspect ratios)
→ Seed2 → Florence-2 grounding/caption → tracks A–F → USER / ASSISTANT.
Tracks
A programmatic QA (counts/style/spatial/absence) from detector
B LLaVA-style (conversation / detailed description / complex reasoning)
C Evol-Instruct (seed atypical Q →… See the full description on the dataset page: https://huggingface.co/datasets/EmpathicRobotics/imagenet-variations-synth-pilot-v3.eval_orange50-variations-2This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.0",
"robot_type": "so100",
"total_episodes": 10,
"total_frames": 2799,
"total_tasks": 1,
"total_videos": 20,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:10"
},
"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/koenvanwijk/eval_orange50-variations-2.Variational-DAPO
Dataset Card for SvS/Variational-DAPO
[🌐 Website] •
[🤗 Dataset] •
[📜 Paper] •
[🐱 GitHub] •
[🐦 Twitter] •
[📕 Rednote]
This dataset consists of 314k variational problems synthesized by the Qwen2.5-32B-Instruct policy during RLVR training on DAPO-17k using the SvS strategy for 600-step training, each accompanied by reference answers.The variational problems undergo a min_hash deduplication with a threshold of 0.85.
Data Loading
from datasets import… See the full description on the dataset page: https://huggingface.co/datasets/RLVR-SvS/Variational-DAPO.Variational-Posterior-PB-7B-GML-mixlerobot_rlbench_variation0_train_open_drawerclean_solar_panel_no_variations_clean_datasetThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 10,
"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/dhirajdg/clean_solar_panel_no_variations_clean_dataset.rollout_smolvla_object_variation_close_curtain_20260919_113226This 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/seriintan/rollout_smolvla_object_variation_close_curtain_20260919_113226.so101_state_machine_variationsrollout_multi_task_dit_object_variation_close_curtain_20260919_122132This 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/seriintan/rollout_multi_task_dit_object_variation_close_curtain_20260919_122132.eval_variation_act_pick_place_tapeThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "so100_follower",
"total_episodes": 5,
"total_frames": 1255,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 500,
"fps": 30,
"splits": {
"train": "0:5"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/aadarshram/eval_variation_act_pick_place_tape.