datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
cub-counterfact
Dataset Card for CounterFact
Of the cmt-benchmark project.
Dataset Details
This dataset is a version of the popular CounterFact dataset, originally proposed by Meng et al. (2022) and re-used in different variants by e.g. Ortu et al. (2024). For this version, the 899 CounterFact samples have been sampled based on the parametric memory of Pythia 6.9B, such that it contains samples for which the top model prediction without context is correct. We note that 546 samples in the… See the full description on the dataset page: https://huggingface.co/datasets/copenlu/cub-counterfact.cub-nq
Dataset Card for NQ
Of the cmt-benchmark project.
Dataset Details
This dataset is a version of the popular NQ dataset, originally proposed by Kwiatkowski et al. (2019). For this version, NQ samples have been obtained based on whether we can recover the gold passage from the original Wikipedia page and for which there is one short answer (less than 5 words in length). The context used in these samples is the correct gold context annotated by the original NQ annotators… See the full description on the dataset page: https://huggingface.co/datasets/copenlu/cub-nq.piperx-put-cube-in-drawer-20260908-87ep
PiperX: put the cube into the drawer — 2026-09-08
87 recorded episodes, 54,463 frames, 30 FPS, approximately 30.26 minutes of recorded frames.
Collected on 2026-09-08 in Asia/Shanghai; source file times span approximately 03:37–05:06.
Task: put the cube into the drawer.
Contents
LeRobot v3.0 layout, dual PiperX arms, 14-dimensional action and state vectors.
Three RGB camera streams (top, left wrist, right wrist) and corresponding depth PNG ZIP archives.
Additional… See the full description on the dataset page: https://huggingface.co/datasets/Travor278/piperx-put-cube-in-drawer-20260908-87ep.cub-druid
Dataset Card for DRUID
Of the cmt-benchmark project.
Dataset Details
This dataset is a version of the DRUID dataset by Hagström et al. (2024). For this version, we have sampled 4,500 DRUID entries for which a "true target" (the factcheck verdict) and a "new target" (the stance of the context) could be found.
Dataset Structure
Thus far, we use two versions of the dataset: gpt2-xl and pythia-6.9b with corresponding validation (200 samples) and test splits… See the full description on the dataset page: https://huggingface.co/datasets/copenlu/cub-druid.so101_pick_cubeDexdata format SO-101 dataset
Dataset Structure
so101_pick_cube
├── videos
│ ├── so101_20260630_103330_filtered
│ │ ├── file-000.mp4_top.mp4
│ │ └── file-000.mp4_wrist.mp4
│ └── ...
└── jsonl
├── episode_00000.jsonl
├── episode_00001.jsonl
└── ...
Task Description
This dataset contains robot manipulation demonstrations for the task:
Task: Pick the cube and place it in the plate
Object variations:… See the full description on the dataset page: https://huggingface.co/datasets/Dexmal/so101_pick_cube.ariel-2025-cube-masked-cacheput_cube_0908_0923_pi05_ensyntava-cube-publication
Genesis Cube or Squid Cube?
A Dual-Use Framework for Artificial Collective Organisms
Version: 1.0Author: John Kalyan BasuAffiliation: Founder, Project SYNTAVA · President, THAL-KI · SwitzerlandPublished: 20 September 2026DOI: 10.5281/zenodo.22854142ORCID: 0009-0009-5561-4441License: CC BY 4.0
Languages
English · Deutsch · 中文 · 日本語 · 한국어 · العربية · Español · हिन्दी · Bahasa Indonesia
The eight translated pages are AI-assisted explanatory summaries… See the full description on the dataset page: https://huggingface.co/datasets/SirJohnBasu/syntava-cube-publication.cube_stack_lerobot_v3
Cube stack — lerobot_v3
Task: stack the red cube on the blue cube
UR7e/GELLO cube-stacking recordings collected on 2026-09-19. Place the red cube on top of the blue cube.
This preview is extracted from take_01_20260919_212809/cam1.mp4; it is not a separate setup photograph or an outcome label.
Contents
60 episodes / 20,177 cam1 master frames at 30 FPS (about 11.2 minutes). Two 1280×720 RGB cameras (cam1 scene, cam2 wrist); no depth. State/action each contain six… See the full description on the dataset page: https://huggingface.co/datasets/Bigenlight/cube_stack_lerobot_v3.ur5e_pick_red_cube_lerobot
UR5e Pick Red Cube (TsFile)
Apache TsFile version of tlpss/ur5e-pick-red-cube.
Overview
A UR5e robot-arm manipulation dataset in the LeRobot v2.1 format. The single task
is to pick up a red cube and place it on a blue square. Each record is one control
frame within a demonstration episode, holding the robot's proprioceptive state,
the commanded action, and per-step reward / success flags.
Scale: 103 episodes, 13,567 frames total (≈ 132 frames per episode on… See the full description on the dataset page: https://huggingface.co/datasets/THULab/ur5e_pick_red_cube_lerobot.rubik-cube-solver-benchmark
Rubik's Cube Solver Benchmark — min2phase Solution Length
Solution length and solve time for the LK Forge Rubik's Cube Solver,
which uses the min2phase (Kociemba two-phase) algorithm. 200 seeded random scrambles per
depth, solving the exact shipped module.
Try it: https://lkforge.com/tools/rubik/
Producer: LK Forge — client-side AI games, solvers, and tools.
Headline numbers (200 cubes each)
Scramble depth (random turns)
Mean solution moves
Median
Max
Mean… See the full description on the dataset page: https://huggingface.co/datasets/LKForge/rubik-cube-solver-benchmark.DreamCubedHumanSample
Dream-Cubed Human Representative Sample
This directory is a deterministic, class-stratified sample of the Dream-Cubed Human dataset. It is provided for reviewer inspection of data quality; the full dataset remains available at https://huggingface.co/datasets/dream-cubed/DreamCubedHumanSample.
The sample dataset for procedurally generated data is available at https://huggingface.co/datasets/dream-cubed/DreamCubedNatural.
Depending on which commands have been run, the sample may… See the full description on the dataset page: https://huggingface.co/datasets/dream-cubed/DreamCubedHumanSample.cubicasa5kclaude-4.5-opus-high-reasoning-250xThis is a reasoning dataset created using Claude Opus 4.5 with a reasoning depth set to high. Some of these questions are from reedmayhew and the rest were generated.
The dataset is meant for creating distilled versions of Claude Opus 4.5 by fine-tuning already existing open-source LLMs.
Stats
Costs: $ 52.3 (USD)
Total tokens (input + output): 2.13 M
Linear-Quad-Cubic_math_dataset
Algebra Equations Dataset
Overview
This dataset contains automatically generated algebra problems with step-by-step solutions.
It is intended for training or testing models that solve equations and show intermediate reasoning steps.
The dataset is stored in JSON Lines (.jsonl) format, where each line represents one problem and its solution.
File
equations_dataset.jsonl
Each line contains:
{
"input": "<problem>",
"output": "<step-by-step solution>"
}… See the full description on the dataset page: https://huggingface.co/datasets/Userpawan/Linear-Quad-Cubic_math_dataset.maverix-fullclaude-sonnet-4.5-high-reasoning-250xThis is a reasoning dataset created using Claude Sonnet 4.5 with a high reasoning effort. Some of these questions are from reedmayhew and the rest were generated.
The dataset is meant for creating distilled versions of Claude Sonnet 4.5 by fine-tuning already existing open-source LLMs.
The default system prompt from OpenrouterAI was used
You are Claude Sonnet 4.5, a large language model from anthropic.
Formatting Rules:
- Use Markdown for lists, tables, and styling.
- Use ```code fences```… See the full description on the dataset page: https://huggingface.co/datasets/cublya/claude-sonnet-4.5-high-reasoning-250x.mem_cube_2This is a MemCube of type memos.configs.mem_cube.GeneralMemCubeConfig.
mem_cube_3This is a MemCube of type memos.configs.mem_cube.GeneralMemCubeConfig.
cubiczan-training-datacuban-spanish-sample
que Cuban Spanish Conversational Sample (v0.3)
A synthetic sample that demonstrates the schema of the que conversational dataset for Cuban Spanish (es-CU). It accompanies the que white paper and shows prospective research partners what a que record looks like. This is synthetic demonstration data, not a collected corpus.
What this is
These records were constructed to the production schema to illustrate its shape. They stand in for data that has yet to be collected… See the full description on the dataset page: https://huggingface.co/datasets/que-app/cuban-spanish-sample.gpt-5.2-high-reasoning-250x
Generated using DataGen by TeichAI
This is a reasoning dataset created using GPT 5.2 with a reasoning depth set to high.
The dataset is meant for creating distilled versions of GPT 5.2 by fine-tuning already existing open-source LLMs.
Stats
Costs: $ 10.58 (USD)
Total tokens (input + output): N\A
dexterous-rdp-cube-data
AutoDL Research Backup
Private migration backup from /root/autodl-tmp. Selection and restore metadata are in SebastianLZJ/autodl-server-archive.
mecharm270_pick_cube_vlm_simplestack-cubes-small-pi05-v1-online-bufferidentity_CUBITsdf-data-cubic_gravityMAVENimt-track-data