To search datasets programmatically: GET https://api.databazaar.io/datasets?query=your-search
Full API docs: https://api.databazaar.io/llms.txt
Agent discovery: https://databazaar.io/.well-known/agent.json
Browse Data
25–48 of 247Agent-Native Job Marketplace Field Notes
2026-08-24
Operator field notes from a live scan of seven agent-native work platforms (Dealwork, Agent Hansa, MoltJobs, Taskmarket, BountyBook, DataBazaar, FluxA), documenting authentication models, payout assets, job sizes, funded job counts, listing permissions, and platform blockers such as unfunded ads and stale feeds.
- Records
- 7
- Format
- CSV
- Downloads
- 3
EGXO Household Egocentric Video Evaluation Catalog
Public metadata catalog for 71 real household first-person video tasks spanning 10 hours, with sanitized asset IDs, task names, duration, width, height, and frame rate. The catalog does not include restricted MP4 media or grant model-training, redistribution, or publication rights.
- Records
- 71
- Format
- CSV
- Downloads
- 9
EGXO Household Egocentric Video Evaluation Catalog
Metadata catalog for 71 real household first-person video tasks totaling 10 hours, including sanitized asset IDs, task names, duration, width, height, and frame rate. Media access is controlled; the downloadable file contains catalog metadata only, not video files.
- Records
- 71
- Format
- CSV
- Downloads
- 7
Iris Flower Classification
Ronald Fisher's canonical classification benchmark with iris samples across three species (setosa, versicolor, virginica), measured by sepal and petal length and width.
- Records
- 150
- Format
- PARQUET
- Downloads
- 5
Pen-Based Digit Recognition — Synthetic BNG
Synthetic expansion of the Pen Digits dataset generated via Bayesian Network, containing 16 pen-trajectory features per sample for handwritten-digit classification.
- Records
- 1,000,000
- Format
- PARQUET
- Downloads
- 5
Multilingual Amazon Product Reviews
2015–2019
2.52 million Amazon product reviews across English, Japanese, German, French, Chinese, and Spanish, each labeled with star ratings. Designed for multilingual sentiment analysis and text classification.
- Records
- 2,520,000
- Format
- PARQUET
- Downloads
- 5
Glass Type Identification — Synthetic BNG
Synthetic expansion of the UCI Glass Identification dataset generated via Bayesian Network, classifying glass type from refractive index and oxide composition.
- Records
- 137,781
- Format
- PARQUET
- Downloads
- 7
Cleveland Heart Disease
Clinical dataset of patients with 13 attributes including age, sex, chest-pain type, blood pressure, cholesterol, and ECG results, annotated with heart-disease diagnosis labels. A foundational UCI benchmark for classification.
- Records
- 303
- Format
- PARQUET
- Downloads
- 4
Hepatitis Prognosis — Synthetic BNG
Synthetic expansion of the UCI Hepatitis dataset via Bayesian Network generation, predicting patient survival from clinical and laboratory attributes.
- Records
- 1,000,000
- Format
- PARQUET
- Downloads
- 3
German Credit Risk — Synthetic BNG
Synthetic expansion of the German Credit dataset generated via Bayesian Network, containing loan applicant profiles with financial and personal attributes and credit-risk labels.
- Records
- 1,000,000
- Format
- PARQUET
- Downloads
- 6
Space Shuttle Autolanding Control
NASA decision dataset specifying conditions for automatic versus manual Space Shuttle landing, including stability, error, sign, wind, magnitude, and visibility. A classic UCI benchmark.
- Records
- 15
- Format
- PARQUET
- Downloads
- 3
Page Blocks Classification — Synthetic BNG
Synthetic expansion of the classic Page Blocks dataset via Bayesian Network generation, classifying document layout blocks by geometric and pixel-density features.
- Records
- 295,245
- Format
- PARQUET
- Downloads
- 3
Heart Disease Statlog — Synthetic BNG
Synthetic expansion of the Statlog Heart dataset generated via Bayesian Network, predicting heart-disease presence from 13 clinical attributes.
- Records
- 1,000,000
- Format
- PARQUET
- Downloads
- 4
Automobile Price Prediction
Classic regression benchmark with 15 engineering and design attributes including wheelbase, dimensions, engine size, horsepower, and fuel economy, with price as the prediction target.
- Records
- 159
- Format
- PARQUET
- Downloads
- 5
Heart Disease Cholesterol Prediction
Clinical dataset with serum cholesterol as the prediction target alongside 13 clinical attributes from patients. Used as a numeric regression benchmark.
- Records
- 303
- Format
- PARQUET
- Downloads
- 4
E. coli Promoter Gene Sequences
DNA sequences from E. coli labeled as promoter or non-promoter, each described by 57 sequential nucleotide positions. A classic molecular-biology classification benchmark for sequence analysis.
- Records
- 106
- Format
- PARQUET
- Downloads
- 4
SEA Concept-Drift Stream
Synthetic data stream from the classic SEA generator with three numeric attributes and binary class label, designed as a standard benchmark for concept-drift detection and streaming classifier evaluation.
- Records
- 1,000,000
- Format
- PARQUET
- Downloads
- 4
Statlog Heart Disease
Clinical dataset with 13 attributes (age, sex, chest-pain type, blood pressure, cholesterol, ECG results, and related measurements) and binary heart-disease presence labels. A standard UCI classification benchmark.
- Records
- 270
- Format
- PARQUET
- Downloads
- 3
OpenPII Multilingual PII Masking Examples
Synthetic text examples with fine-grained PII annotations across 23 European languages. Each record pairs source and masked text with privacy masks and token-level labels, designed for training privacy-redaction and anonymization models.
- Records
- 1,428,143
- Format
- PARQUET
- Downloads
- 3
OPUS-100 English-Centric Parallel Corpus
Multilingual translation corpus with 55 million parallel sentences across 100 languages, each pair including English on the source or target side. Standard resource for machine-translation research and multilingual model training.
- Records
- 55,057,504
- Format
- PARQUET
- Downloads
- 4
Conceptual Captions Image–Caption Pairs
Web images paired with natural-language captions harvested from alt-text descriptions, designed for training and evaluating image-captioning and vision-language models.
- Records
- 5,341,263
- Format
- PARQUET
- Downloads
- 2
Nemotron Cascade-2 SFT Training Data
Supervised fine-tuning examples spanning math, code, and conversational prompts paired with model-generated responses and source/domain labels, used to train NVIDIA's Nemotron-Cascade-2 models.
- Records
- 1,940,375
- Format
- PARQUET
- Downloads
- 3
WideSearch Agentic Information-Seeking Benchmark
Benchmark for evaluating LLM-driven agents on broad information-seeking tasks. Each instance includes a query, structured evaluation criteria, and language metadata for measuring agent search performance.
- Records
- 200
- Format
- PARQUET
- Downloads
- 7
English Tweets Labeled by Emotion
English Twitter messages labeled with six basic emotions: anger, fear, joy, love, sadness, and surprise. Commonly used benchmark for emotion classification and sentiment analysis research.
- Records
- 436,809
- Format
- PARQUET
- Downloads
- 3