datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
arc_agi_v2_public_evalarc-agi-2arc-agi-2-reasoning-5
ARC AGI 2 Reasoning Traces
A collection of reasoning traces for writing python programs that solve ARC AGI II tasks.
All rows in this data involve correct programs - correct on both train and test examples.
Tasks are drawn from the ARC AGI I and ARC AGI training sets.
Data sources include:
Programs taken from the SOAR dataset. Note that those only include ARC AGI I tasks.
Programs generated using private API endpoints, including GPT-OSS, GPT-5-mini, Qwen-2.5 and others. Models are… See the full description on the dataset page: https://huggingface.co/datasets/Trelis/arc-agi-2-reasoning-5.Harbor-Parity-Test-ARC-AGI-2arc_agi_v2
ARC-AGI-2 Dataset (A Take On Format)
This dataset is a reorganized version of the ARC-AGI-2 (Abstraction and Reasoning Corpus for Artificial General Intelligence v2) benchmark, formatted for HuggingFace Datasets.
Dataset Structure
The original ARC-AGI-2 dataset has been transformed from its file-based JSON structure into a standardized HuggingFace dataset with two splits:
train (1000 examples): Tasks from the original training directory
test (120 examples): Tasks from… See the full description on the dataset page: https://huggingface.co/datasets/Ardea/arc_agi_v2.arc_agi_2_human_testing
ARC-AGI-2 Human testing data
This file contains data from human testing sessions on ARC-AGI tasks.
Each row represents a single test attempt by a human participant on a specific task-test pair in the "Public Train" or "Public Eval" ARC-AGI-2 datasets. Not all tasks in the released "Public Train"
sets were tested, so these results are not comprehensive. This data does not include tasks from "Semi Private Evaluation" or "Private Evaluation"
Column Descriptions… See the full description on the dataset page: https://huggingface.co/datasets/arcprize/arc_agi_2_human_testing.arc-agi-2
ARC-AGI-2 Few-Shot Conversations
ARC-AGI-2 is a benchmark of 1,000 public training tasks and 120 public evaluation tasks for assessing reasoning systems. This repository packages the public tasks into a Hugging Face–friendly format with:
canonical/original arc-agi 2 train/evaluation splits
Parquet shards for fast downloads & streaming
per-example PNG renderings of every grid (demonstration and test)
text prompts & full conversations ready for LLM fine-tuning
Dataset… See the full description on the dataset page: https://huggingface.co/datasets/vincentkoc/arc-agi-2.arc-agi-2-evaluationARC-AGI-2The ARC-AGI-2 dataset downloaded from the ARC-AGI-2 kaggle competition. This dataset has 1000 training tasks and 112 public eval tasks.
Note 1: Most other ARC-AGI-2 datasets on HuggingFace are missing the answers to the eval set even though the ARC-AGI-2 creators released it. We make sure to include the answers to the eval set here.
Note 2: Technically, ARC-AGI-2 has four splits: train, public eval, semi-private eval, private eval. The train and public eval sets are here. But the semi-private… See the full description on the dataset page: https://huggingface.co/datasets/eturok/ARC-AGI-2.arc-agi-2-all-100
100k Rows of Python Programs to solve ARC AGI I and II Tasks
Rows include programs that:
Get all examples (train and test) correct.
Get at least one train or test example correct.
Get no train or test examples correct BUT do produce valid output grids (i.e. 30x30 or less).
Tasks are drawn from the ARC AGI I and ARC AGI training sets.
Data sources include:
- Programs taken from the SOAR dataset. Note that those only include ARC AGI I tasks.
- Programs generated using… See the full description on the dataset page: https://huggingface.co/datasets/Trelis/arc-agi-2-all-100.arc-agi-2-trainingarc-agi-2-grids
ARC-AGI-2 Grids — training + analysis corpus (NVARC-compatible)
Companion dataset for the Kaggle ARC Prize 2026 (ARC-AGI-2) solver built on
sorokin/qwen3_4b_grids15_sft139 + per-task rank-256 LoRA (NVARC lineage).
Everything here is generated from public canonical data only (1,000
training / 120 evaluation tasks); no hidden competition data is included.
Contents
Path
Rows
Description
train/train_tasks.jsonl
1,000
canonical training tasks (full I/O)… See the full description on the dataset page: https://huggingface.co/datasets/Nabidnur/arc-agi-2-grids.arc-agi-2Dataset from 'https://github.com/arcprize/ARC-AGI-2/commits/main/'
Commit version f3283f7 (May 16, 2025)
Note that github dataset has 1000 examples in train and 120 in test. But because some files contains several tests they were flattened.
arc-agi-2
Description
ARC-AGI 2 dataset reformatted for easy fine-tuning.
RPS-ARC-AGI-1-and-2This is the DPO dataset used in the RPS paper ( https://github.com/iamjasonfeng/RPS-Paper )
This dataset is based on the following dataset from Trelis: https://huggingface.co/datasets/Trelis/arc-agi-2-reasoning-5
arc-agi-2-cot-sft
ARC-AGI-2 CoT-Solving SFT Dataset
Companion to Nabidnur/arc-agi-2-grids.
Grid-based chain-of-thought transcripts for the ARC-AGI-2 Kaggle competition,
NVARC format compatible (Qwen chat template, {" "}-separated digit rows).
Tiers
program-verified — a transformation rule fitted ONLY on the demonstrations
reproduces all of them exactly (verified=true). Hypothesis + per-pair
verification + application.
program-true (synthetic) — rules from generator ancestry… See the full description on the dataset page: https://huggingface.co/datasets/Nabidnur/arc-agi-2-cot-sft.arc-agi-2-vision
ARC-AGI-2 Vision Dataset
Image-rendered ARC-AGI-2 grids for vision-based solvers (e.g. Qwen-VL class
models). Companion to Nabidnur/arc-agi-2-grids
(text/grid SFT) and Nabidnur/arc-agi-2-cot-sft
(reasoning transcripts).
Render spec
Standard ARC 10-color palette (see meta/manifest.json).
White 4px margin, 1px dark grid lines, cell scale = max(8, 448//max_dim) px.
image column: PNG bytes (HF Image-compatible {bytes, path} struct).
Contents… See the full description on the dataset page: https://huggingface.co/datasets/Nabidnur/arc-agi-2-vision.arc-agi-2-partialplus-max
~40k Rows of Python Programs to solve ARC AGI I and II Tasks
Rows include programs that:
Get all examples (train and test) correct.
Get at least one train or test example correct.
Tasks are drawn from the ARC AGI I and ARC AGI training sets.
Data sources include:
Programs taken from the SOAR dataset. Note that those only include ARC AGI I tasks.
Programs generated using private API endpoints, including GPT-OSS, GPT-5-mini, Qwen-2.5 and others. Models are listed in a dataset… See the full description on the dataset page: https://huggingface.co/datasets/Trelis/arc-agi-2-partialplus-max.arcagi2plan-crl-arc-agi-2
PLAN-CRL ARC-AGI-2
This is a Hugging Face mirror/packaging of the public ARC-AGI-2 tasks used by the PLAN-CRL ARC-AGI-2 evaluation harness.
Source: https://github.com/arcprize/ARC-AGI-2
Splits
training: 1000 tasks
evaluation: 120 tasks
Columns
task_id: upstream ARC task id.
train: ARC training examples with input and output grids.
test_inputs: test input grids only.
test_outputs: reference output grids used by the evaluator.
prompt_json: compact JSON shown… See the full description on the dataset page: https://huggingface.co/datasets/1337xyz1337xyz/plan-crl-arc-agi-2.arc-agi-2-synthetic-v1
ARC-AGI-2 Synthetic Curriculum v1
Program-generated ARC-like tasks with full ancestry records and
independent verification (every task's rule must be re-fittable from its
own demonstrations; 44 degenerate tasks rejected).
Families (7): d4_transform, color_remap, tile_k, crop_bbox, symmetry_fill,
hole_recolor, path_propagation.
Counts
family
verified
d4_transform
450
color_remap
450
tile_k
450
crop_bbox
450
symmetry_fill
439
hole_recolor
417… See the full description on the dataset page: https://huggingface.co/datasets/Nabidnur/arc-agi-2-synthetic-v1.ARC-AGI-2arc-agi-1-kimi-k2ARC-AGI-2_training_flattenedarc-agi-2-evalarc_agi_2_augmentedarc_agi_v2arc-agi-2-partial-100Original-ARC-AGI-2arc-agi-2-train