1,002 datasets

Hugging Face Datasets2026 · Table · Parquet · gated

KRS scrape every entry in Poland's National Court Register (Krajowy Rejestr Sądowy), about 1.2 million companies and associations, parsed from the full extract PDFs and updated weekly. from datasets import load_dataset krs = load_dataset("tiagozip/krs", split="train") format one parquet file, data/krs.parquet, one row per KRS number. column type notes krs string 10 digit KRS number register string

Hugging Face Datasets2026 · Text · gated

Meddies ASR Bench This public dataset repository contains the validated 10 full audio chunks used for the MOSS + Gemini 3.8 Flash smoke benchmark. Contents data/full_10_chunks/audio/: ten full MP3 input chunks. data/full_10_chunks/moss/: the ten source MOSS-Diarize-Transcribe JSON records. data/full_10_chunks/manifest.json: chunk paths, offsets, durations, and embedded MOSS text. results/: the dow

Hugging Face Datasets2026 · dataset · gated

AIGC Dataset — NovelAI Generations ⚠️ Contains NSFW content. Images and prompts have not been filtered. Many are sexually explicit. 1,562,685 AI-generated images (≈332 GB) made with NovelAI's image models (V3 → V4 → V4.5 → V5) between 2023-11 and 2026-09 through a private Discord bot (Kohaku-NAI). Every image has its full generation metadata (model, sampler, steps, CFG, seed, prompt, …) and an exa

Hugging Face Datasets2026 · dataset · gated

Dynovia Crash Open 200 vehicle crash simulation cases across sedans, an SUV, and pickup trucks, for building and testing crash prediction models. Dynovia Crash Open gives engineers and researchers simulation data to develop, train, and evaluate their own models for non-commercial research. It covers five vehicle models, with multiple design configurations for each vehicle to capture variation in c

Hugging Face Datasets2026 · dataset · gated

Hugging Face Datasets2026 · Text · gated

PathoVernier PathoVernier is a benchmark for quantitative cell-composition reasoning on H&E histopathology patches. Each question requires counting specific nucleus types in specific image regions and deriving an answer from those counts with a deterministic rule. Reference counts come from expert nucleus annotations, so a model's reported counts can be checked in addition to its final answer. 759

Hugging Face Datasets2026 · dataset · gated

Counter-GEO-Bench Counter-GEO-Bench evaluates defenses against information-distorting generative engine optimization (GEO) in retrieval-augmented and generative search systems. The gated release covers 247 queries. It includes all 247 information-preserving (IP) rewrites and 237 information-distorting (ID) rewrites. Ten potentially sensitive ID rewrites are withheld; researchers may request access

Hugging Face Datasets2026 · dataset · gated

Hugging Face Datasets2026 · Magnetic resonance imaging · gated

Dataset Card for OAI Knee MRI (Preprocessed 3D NumPy Arrays) Dataset Details Dataset Description This dataset contains 92,313 verified, preprocessed Knee MRI volumes from the Osteoarthritis Initiative (OAI). The original DICOM/medical imaging files have been standardized, cleaned, and converted into highly efficient 3D .npy (NumPy) arrays to accelerate machine learning and computer vision workflow

Hugging Face Datasets2026 · Table · Parquet · gated

Hugging Face Datasets2026 · dataset · gated

OpenAI–Hugging Face Incident Agent Traces Dataset Details OpenAI–Hugging Face Incident Agent Traces This dataset contains redacted agent traces from the OpenAI–Hugging Face security incident, released publicly with permission. The records document agent activity during cybersecurity evaluation runs, including model interactions and, where present, tool calls, outputs, timestamps, and coordination

Hugging Face Datasets2026 · Table · Parquet · gated

Hugging Face Datasets2026 · dataset · gated

Hugging Face Datasets2026 · dataset · gated

Hugging Face Datasets2026 · Image · gated

Car Damage Images A raw image collection for vehicle damage assessment. Unlabeled: these images have no annotations yet and are intended as source material for labeling or pre-training. Structure images/ 001/ images_001.jpg images_002.jpg ... 002/ ... ... 683 folders thumbnails/ 001/ thumbnail_001.jpg thumbnail_002.jpg ... ... 683 folders Branch Files Folders Size… See the full description on the

Hugging Face Datasets2026 · Text · gated

Hugging Face Datasets2026 · Image · gated

Tidy Up: U.S. kitchens before and after cleaning Real kitchens. Real clutter. More than photos. Explore 20 U.S. kitchens before and after cleaning, with photos, LiDAR scans, depth data, object annotations, and generated 3D reconstructions. Study how everyday spaces change—and connect what’s visible in an image to the geometry of the room. A practical sample for exploring scene understanding, 3D re

Hugging Face Datasets2026 · Text · gated

Dissei Financial Judgment — Full Evaluation Package Harbor · GitHub · Website · Hugging Face Dissei explores financial reasoning and analysis behind institutional investment and credit decisions: interpreting evidence, weighing trade-offs and reaching well-supported conclusions. Financial Judgment contains seven tasks drawn from one completed private-equity deal. Each asks for a written analytical

Hugging Face Datasets2026 · Text · gated

k9x4-store Cached intermediate files (numpy arrays and parquet tables) for a private project. Downloads require approval. Paths mirror the project repository's tree. From a checkout of that repository, restore files with: python scripts/offload_cache.py pull --paths 'results/<run>/<file-glob>'

Hugging Face Datasets2026 · Table · CSV · gated

Tiberian Biblical Hebrew → IPA (BHS-aligned) Org: tiberianaiAuthors: John Locke, Teodor BorsCompanion model: tiberianai/tiberian-hebrew-ipa-byt5 Verse-level parallel corpus for distilling Masoretic Biblical Hebrew (niqqud and teʿamim) into Tiberian IPA, using Geoffrey Khan’s reconstruction of the Tiberian pronunciation tradition as encoded by our rule-based teacher. Gated (manual approval). The ca

Hugging Face Datasets2026 · dataset · gated

DexCanvas DexCanvas: Bridging Human Demonstrations and Robot Learning for Dexterous Manipulation Project page · Paper (arXiv:2510.15786) · Repository DexCanvas is a dataset of 111,857 complete right-hand trajectories and 55,442,683 frames of rigid hand–object interaction. Every Parquet row is one trajectory with a shared time axis for hand kinematics, object pose, and per-frame contacts. For each

Hugging Face Datasets2026 · dataset · gated

Atlantis Atlantis is a projected worked out through ECMWF's 2026 CodeForEarth competitionTeam members where: Kalfas Ioannis & Lagaras Stylianos Mentors: Calum Baugh Gianpaolo Balsamo Kenza Tazi Andreas Grafberger Github repository Atlantis Archive is produced with our project's repository: https://github.com/opageo/atlantis Specifications The Archive is a Zarr V3 using shards of (32, 256, 256) dim

Hugging Face Datasets2026 · Text · gated

RLCDAlignBench Paper: Just Ask Jev: Reinforcement Learning for Calibrated Decisions as a Zero-Shot Detector of AI Alignment Failures (arXiv:2609.29429) Code: github.com/sumleo/RLCDAlignBench · Project page: sumleo.github.io/RLCDAlignBench RLCDAlignBench measures whether a detector can tell when a language model's output is an alignment failure. It has 44 benchmarks across ten failure types and fiv

Hugging Face Datasets2026 · Table · Parquet · gated

Amazon ML Challenge 2026: Business Entity Resolution Dataset This repository hosts the official dataset for the Amazon ML Challenge 2026 — Business Entity Resolution Challenge, packaged for high-speed multi-server distributed training, EDA, and evaluation. Both the original TSV files and optimized Parquet files (zstd-compressed) are provided. 📊 Dataset Overview & Statistics Total records across tr

Hugging Face Datasets2026 · dataset · gated