datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
XVLA-Soft-Fold
Cloth-Folding Dataset for X-VLA Paper
This dataset contains 1,500 episodes of cloth folding, collected using Agilex's robotic arm. It was used in the X-VLA paper for cloth-folding tasks, showcasing a near-perfect success rate in folding accuracy.
Dataset Overview
Total Episodes: ～1,500
Task: Automated cloth folding
Robot: Agilex Aloha
Performance: Near 100% success rate in completing the folding task
Hardware setup
We observed that the camera setup of the… See the full description on the dataset page: https://huggingface.co/datasets/Facebear/XVLA-Soft-Fold.Libero-XVLA-formataction-atlas-xvla-saetrainXVLA-Soft-Fold
Cloth-Folding Dataset for X-VLA Paper
This dataset contains 1,500 episodes of cloth folding, collected using Agilex's robotic arm. It was used in the X-VLA paper for cloth-folding tasks, showcasing a near-perfect success rate in folding accuracy.
Dataset Overview
Total Episodes: ～1,500
Task: Automated cloth folding
Robot: Agilex Aloha
Performance: Near 100% success rate in completing the folding task
Hardware setup
We observed that the camera… See the full description on the dataset page: https://huggingface.co/datasets/gop0429/XVLA-Soft-Fold.xvla-soft-foldThis dataset was created using LeRobot.
Dataset Description
Repository: X-VLA
License: Apache 2.0
Paper: Zheng et al., 2025, “X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model” (arXiv:2510.10274)
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "franka",
"total_episodes": 1542,
"total_frames": 2852512,
"total_tasks": 1,
"chunks_size": 1000… See the full description on the dataset page: https://huggingface.co/datasets/lerobot/xvla-soft-fold.lerobot-xvla-soft-fold
LeRobot X-VLA Soft-Fold (Lance Format)
A Lance-formatted version of lerobot/xvla-soft-fold — a multi-camera robotics dataset from the X-VLA project — packaged as three Lance tables for efficient frame-level training, episode-level trajectory loading, and direct access to the original encoded videos. Available directly from the Hub at hf://datasets/lance-format/lerobot-xvla-soft-fold/data.
1,542 episodes
2,852,512 frames at 20 FPS
3 camera streams per episode — cam_high… See the full description on the dataset page: https://huggingface.co/datasets/lance-format/lerobot-xvla-soft-fold.libero_XVLA_datasetaction-atlas-xvla
Action Atlas: X-VLA sparse autoencoders and concepts
Sparse autoencoders (SAEs) and identified concepts for X-VLA, part of the Action Atlas
release accompanying the paper on cross-task activation injection in vision-language-action models.
The interactive explorer is at https://action-atlas.com.
What is here
saes/ TopK SAEs (k=64, 8x expansion) over the X-VLA (soft-prompted Florence-Large), 24 layers, 1024-dim.
Arms present: per-token 48, mean-pool 48. Per-token… See the full description on the dataset page: https://huggingface.co/datasets/bag100/action-atlas-xvla.eval_xvla_pegboard_in-distributionThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_follower",
"total_episodes": 52,
"total_frames": 30507,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:52"
},
"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/allenai/eval_xvla_pegboard_in-distribution.eval_xvla-so101-4tasks-aug_stack_25This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "so_follower",
"total_episodes": 20,
"total_frames": 49728,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:20"
},
"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/hjkso1406/eval_xvla-so101-4tasks-aug_stack_25.eval_xvla-so101-4tasks-aug_sort_25This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "so_follower",
"total_episodes": 20,
"total_frames": 61173,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:20"
},
"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/hjkso1406/eval_xvla-so101-4tasks-aug_sort_25.eval_xvla_candy_sorting_in-distributionThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_follower",
"total_episodes": 50,
"total_frames": 33307,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:50"
},
"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/allenai/eval_xvla_candy_sorting_in-distribution.eval_xvla_cup_stacking_in-distributionThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_follower",
"total_episodes": 50,
"total_frames": 31683,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:50"
},
"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/allenai/eval_xvla_cup_stacking_in-distribution.eval_xvla_linearbot_in-distributionThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_linear_bot",
"total_episodes": 50,
"total_frames": 31908,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:50"
},
"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/allenai/eval_xvla_linearbot_in-distribution.eval_xvla-so101-4tasks-aug_pick-place_25This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "so_follower",
"total_episodes": 20,
"total_frames": 36058,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:20"
},
"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/hjkso1406/eval_xvla-so101-4tasks-aug_pick-place_25.finetune_xvla
🤖 X-VLA: Soft-Prompted Transformer as a Scalable Cross-Embodiment Vision-Language-Action Model
📄 Paper
🌐 Project Page
🤗 Hugging Face
Read the Full Research
Explore the Demos
Access Models & Datasets
🏆 Highlights & News
🎉 Exciting News: X-VLA Accepted to ICLR 2026
We are thrilled to announce that X-VLA has been accepted to ICLR 2026.
🚀 Now Supported in LeRobot
X-VLA is now natively integrated into the LeRobot platform.Give it… See the full description on the dataset page: https://huggingface.co/datasets/yqi19/finetune_xvla.eval_xvla_toy_storing_in-distributionThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_follower",
"total_episodes": 50,
"total_frames": 34854,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:50"
},
"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/allenai/eval_xvla_toy_storing_in-distribution.xvla-pant-short-success-collecteval_xvla_pegboard_in-distributionThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_follower",
"total_episodes": 52,
"total_frames": 30507,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:52"
},
"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/Jiafei1224/eval_xvla_pegboard_in-distribution.eval_xvla_candy_sorting_in-distributionThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_follower",
"total_episodes": 50,
"total_frames": 33307,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:50"
},
"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/Jiafei1224/eval_xvla_candy_sorting_in-distribution.eval_xvla_toy_storing_in-distributionThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_follower",
"total_episodes": 50,
"total_frames": 34854,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:50"
},
"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/Jiafei1224/eval_xvla_toy_storing_in-distribution.eval_xvla_linearbot_in-distributionThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_linear_bot",
"total_episodes": 50,
"total_frames": 31908,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:50"
},
"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/Jiafei1224/eval_xvla_linearbot_in-distribution.eval_xvla_cup_stacking_in-distributionThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_follower",
"total_episodes": 50,
"total_frames": 31683,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:50"
},
"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/Jiafei1224/eval_xvla_cup_stacking_in-distribution.eval_xvla_pipette_oodThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_follower",
"total_episodes": 23,
"total_frames": 11466,
"total_tasks": 3,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:23"
},
"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/allenai/eval_xvla_pipette_ood.eval_xvla_cup_stacking_oodThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_follower",
"total_episodes": 25,
"total_frames": 16096,
"total_tasks": 3,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:25"
},
"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/allenai/eval_xvla_cup_stacking_ood.eval_xvla_candy_sorting_oodThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_follower",
"total_episodes": 23,
"total_frames": 16014,
"total_tasks": 4,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:23"
},
"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/allenai/eval_xvla_candy_sorting_ood.eval_xvla_test_tube_oodThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_follower",
"total_episodes": 24,
"total_frames": 15020,
"total_tasks": 3,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:24"
},
"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/allenai/eval_xvla_test_tube_ood.action-atlas-xvla-libero-activationsrollout_eval_xvla_frazier_20260907_104650This 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_eval_xvla_frazier_20260907_104650.eval_xvla_pipette_in-distributionThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "bi_yam_follower",
"total_episodes": 50,
"total_frames": 24471,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:50"
},
"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/allenai/eval_xvla_pipette_in-distribution.