datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
ck3-gameplay-mouse-keyboard-dataset
👑 CRUSADER KINGS III: HUMAN GAMEPLAY MULTIMODAL TELEMETRY DATASET
🏰 The Sovereign Grand Strategy Benchmark for Vision-Language-Action (VLA) Foundation Models & Autonomous GUI Agents
💬 "Every click. Every drag. Every panicked Alt-Tab before an unexpected inheritance crisis."
📍 Table of Contents
🏰 Executive Summary & Research Vision… See the full description on the dataset page: https://huggingface.co/datasets/Ethosoft/ck3-gameplay-mouse-keyboard-dataset.agilex_tian_open_laptop_press_keyboardThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.0",
"robot_type": "arx5_bimanual",
"total_episodes": 20,
"total_frames": 15415,
"total_tasks": 1,
"total_videos": 60,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 25,
"splits": {
"train": "0:20"},
"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/villekuosmanen/agilex_tian_open_laptop_press_keyboard.so101-keyboard-typing-benchmark
SO-101 Keyboard Typing Checkpoints
Reproducibility bundle for the SO-101 six-letter keyboard-typing comparisons.
The policies use a fixed Cartesian keyboard map, proprioceptive joint state,
registered keypress feedback, and no vision or online pose correction.
Physical typing example — September 17, 2026
The newly calibrated scene and model_2498.pt completed one physical NEWTON
trial in 7.099 s, after 500 P1A + 2,000 Transit15 updates and 2,048/2,048 strict… See the full description on the dataset page: https://huggingface.co/datasets/VigneshBhavan/so101-keyboard-typing-benchmark.pusht_physical_keyboard_15fpsThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "nova",
"total_episodes": 90,
"total_frames": 80168,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 15,
"splits": {
"train": "0:90"
},
"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/StefanWagnerWandelbots/pusht_physical_keyboard_15fps.2018-Away-from-Keyboard-Transcripts
2018 Away from Keyboard Transcripts
Complete transcripts from the 2018 episodes of the Away from Keyboard podcast.
Generated from this GitHub repository.
Hypa-Keyboard-v1
Hypa Keyboard v1 is a 409,598-example instruction dataset for training on-device smart keyboard models — next-word prediction, word completion, autocorrect, and grammatical error correction — across 27 languages, weighted heavily toward African languages that no mainstream keyboard supports.
Every example is a three-turn chat (system → user → assistant), so the dataset can be fed directly to any chat-template SFT pipeline. Errors in the input text are synthetically injected from a… See the full description on the dataset page: https://huggingface.co/datasets/hypaai/Hypa-Keyboard-v1.keyboards
Description
https://github.com/makcedward/nlpaug/tree/master/nlpaug/res/char/keyboard
Citation
@misc{ma2019nlpaug,
title={NLP Augmentation},
author={Edward Ma},
howpublished={https://github.com/makcedward/nlpaug},
year={2019}
}
Hypa-Keyboard-v2
Hypa Keyboard v2 is a 409,598-example instruction dataset for training on-device smart keyboard models — next-word prediction, word completion, autocorrect, and grammatical error correction — across 27 languages, weighted heavily toward African languages that no mainstream keyboard supports.
Every example is a three-turn chat (system → user → assistant), so the dataset can be fed directly to any chat-template SFT pipeline. Errors in the input text are synthetically injected from a… See the full description on the dataset page: https://huggingface.co/datasets/hypaai/Hypa-Keyboard-v2.gym_hil_pickplace_bin_keyboardThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 10,
"features": {
"action": {
"dtype": "float32",
"shape": [
4
],
"names": [
"delta_x",
"delta_y",
"delta_z",
"gripper"
]
},
"next.reward": {
"dtype": "float32"… See the full description on the dataset page: https://huggingface.co/datasets/hungdo2401/gym_hil_pickplace_bin_keyboard.franka_vla_keyboard_joint_delta
Franka VLA keyboard lift-block demonstrations
This private LeRobot v3 dataset contains keyboard-teleoperated Franka lift-block
demonstrations prepared for fine-tuning
DAVIAN-Robotics/pi05_droid_jointpos.
Dataset summary
98 episodes and 24,831 action-aligned frames at 20 Hz
Task text: Lift the block.
observation.images.base_0_rgb: exterior camera, 224 x 224 RGB
observation.images.left_wrist_0_rgb: wrist camera, 224 x 224 RGB
observation.state: 8 values [q0..q6… See the full description on the dataset page: https://huggingface.co/datasets/Ameyapores/franka_vla_keyboard_joint_delta.asleep_keyboardThe Asleep at the Keyboard dataset contains 89 code generation scenarios that are designed to test the ability of code generation models to generate code secure code. The dataset is split into three evaluation axes: diversity of weaknesses (DoW), diversity of prompts (DoP), and diversity of domains (DoD).
To perform this analysis we prompt Copilot to generate code in scenarios relevant to high-risk cybersecurity weaknesses, e.g. those from MITRE’s “Top 25” Common Weakness Enumeration (CWE) list. We explore Copilot’s performance on three distinct code generation axes—examining how it performs given diversity of weaknesses, diversity of prompts, and diversity of domains. In total, we produce 89 different scenariospb-hobi-keyboard-typing-v1
PB Hobi Keyboard Typing V1
This dataset is part of the IMBench suite,
distributed in the LeRobot v3.0 format.
Overview
Episodes
10
Total frames
25611
FPS
100
Robot
Panda
Cameras
4 (rgb_agent_cam, rgb_exo_bird_cam, rgb_gripper0_cam, rgb_gripper1_cam)
Visualize
Browse episodes interactively in the LeRobot Visualizer:
→ https://lerobot-visualize-dataset.hf.space/imbench/pb-hobi-keyboard-typing-v1/episode_0
Load with LeRobot… See the full description on the dataset page: https://huggingface.co/datasets/imbench/pb-hobi-keyboard-typing-v1.keyboard-key-coverage-20260909
Keyboard key coverage preview media
407 continuous 15-second derived game clips and allowlisted keyboard/action records.
Browse the evidence and limitations at https://mikusama99-keyboard-key-coverage-20260909.static.hf.space/
Per-case original repositories, pinned revisions, source hashes and input timestamps are recorded in annotations. D2E is excluded. Game copyrights and upstream licenses apply; this is an automated evidence preview, not manually certified training data. No… See the full description on the dataset page: https://huggingface.co/datasets/mikusama99/keyboard-key-coverage-20260909.keyboard_pick_cube_30_episodesomx_keyboard_dataset
Task_150_keyboard_2_MCAP
Created with Cyclo Intelligence by ROBOTIS.
German-KeyboardLM-Corpus
German KeyboardLM Training Corpus
Dieser Datensatz wurde für das Training eines ultrakompakten 33M-Parameter-Sprachmodells für mobile On-Device-Tastaturen (FUTO Keyboard) zusammengestellt.
Datenquellen & Herkunft
Der Korpus ist eine kuratierte Zusammenstellung aus folgenden Open-Source-Datensätzen:
German Wikipedia Dumps (CC BY-SA 4.0)
OpenAssistant Conversations (OASST) (Apache 2.0)
Leipzig Corpora Collection / News (CC BY)
Durchgeführte… See the full description on the dataset page: https://huggingface.co/datasets/VerbalJungle/German-KeyboardLM-Corpus.pick_cube_keyboardWaddle_xlerobot_keyboard_v2This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "xlerobot",
"total_episodes": 1,
"total_frames": 2990,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 500,
"fps": 30,
"splits": {
"train": "0:1"
},
"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/DozenDucc/Waddle_xlerobot_keyboard_v2.so100_keyboard_teleopThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "so_follower",
"total_episodes": 2,
"total_frames": 2521,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:2"
},
"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/Sri-Ram-A/so100_keyboard_teleop.43-143-phase2-appconv-ime-keyboard-noisekeyboard_record_1
keyboard_record_1
This dataset was generated using phosphobot.
This dataset contains a series of episodes recorded with a robot and multiple cameras. It can be directly used to train a policy using imitation learning. It's compatible with LeRobot.
To get started in robotics, get your own phospho starter pack..
Desksetup-keyboard-mouse-editing-v1
Desksetup-keyboard-mouse-editing-v1
Description
A paired image-editing dataset for desk setups. Each sample includes an instruction prompt, a base setup image, a product reference image (keyboard or mouse), and a target edited image showing realistic product placement/replacement.
Splits
keyboard
mouse
Features
sample_id: string
item_key: string
prompt: string
base_image: Image
ref_image: Image
target_image: Image
Source
Repo id:… See the full description on the dataset page: https://huggingface.co/datasets/EmreAkgul/Desksetup-keyboard-mouse-editing-v1.il_gym_keyboard_100This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": null,
"total_episodes": 100,
"total_frames": 7209,
"total_tasks": 1,
"total_videos": 200,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 10,
"splits": {
"train": "0:100"},
"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/swagatk/il_gym_keyboard_100.ur5e_keyboardThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "ur5e",
"total_episodes": 1,
"total_frames": 71,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 500,
"fps": 15,
"splits": {
"train": "0:1"
},
"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/scylearning/ur5e_keyboard.sim_keyboard_datasetThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": null,
"total_episodes": 30,
"total_frames": 1283,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 10,
"splits": {
"train": "0:30"
},
"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/lisirui333/sim_keyboard_dataset.Keyboardsim_keyboard_dataset3This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": null,
"total_episodes": 30,
"total_frames": 1367,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 10,
"splits": {
"train": "0:30"
},
"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/alexanta/sim_keyboard_dataset3.sim_keyboard_dataset2This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": null,
"total_episodes": 30,
"total_frames": 2861,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 10,
"splits": {
"train": "0:30"
},
"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/alexanta/sim_keyboard_dataset2.eval_pusht_keyboard_policy_100This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "nova",
"total_episodes": 1,
"total_frames": 6288,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:1"
},
"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/StefanWagnerWandelbots/eval_pusht_keyboard_policy_100.imnet1k_typewriter_keyboard