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

1–24 of 104
Filters
Newest
text
$49

Enterprise MCP & AI Agent Function-Calling Policy Benchmark v1 (3,000-Row CSV)

Updated Nov 16, 2026

Curated by Oxford Digital (oxforddigitals.com). 3,000 schema-validated JSON-RPC 2.0 Model Context Protocol (MCP) and AI Agent tool-call policy evaluation & fine-tuning records across 8 enterprise verticals (FinTech & Billing, Cloud IAM, DevOps CI/CD, Healthcare FHIR, Enterprise CRM, SQL Analytics, Workspace Calendar/Mail, and Kubernetes Ops). Balanced across 5 ground-truth policy outcomes (600 rows each: ALLOW_AUTO, REQUIRE_HITL_APPROVAL, DENY_SCOPE_MISMATCH, DENY_TENANT_ISOLATION, DENY_QUOTA_OR_BUDGET_CAP) with paired Python Pydantic v2 and TypeScript Zod deterministic guardrail validators for SFT, DPO, GRPO, and policy-router Precision/Recall benchmarking.

Records
3,000
Format
CSV
Downloads
0
Sample structure100/100
text
$99

Code & AI Security Remediation Suite v4 — Single-Seat Indie Edition (6,000-Row CSV)

Updated Nov 15, 2026

Single-Seat Indie & Researcher commercial license curated by Oxford Digital (oxforddigitals.com). Contains all 6,000 unseen defensive code & AI security audit-to-remediation pairs in structured UTF-8 CSV format across 5 stacks: (1) 1,200 Solidity v0.8.20-0.8.28 EVM smart contracts, (2) 1,200 Rust/Solana Anchor programs, (3) 1,200 Python/TypeScript Autonomous AI Agent, MCP, RAG & RLVR guardrail modules, (4) 1,200 Go/TypeScript/Python Cloud & FinTech API services, and (5) 1,200 Kubernetes, Terraform & CI/CD IaC configurations. Zero overlap with public Hugging Face v1-v3 splits. Licensed for 1 individual developer or security researcher seat (for multi-seat Team Pro Parquet+CSV+JSONL archive, see our $599 Enterprise Suite).

Records
6,000
Format
CSV
Downloads
0
Sample structure100/100
text
$599

Enterprise Multi-Domain Code & AI Security Remediation Suite v4 (6,000-Row Full Archive)

Updated Nov 13, 2026

Curated by Oxford Digital (oxforddigitals.com). 6,000 unseen defensive security remediation pairs across 5 high-demand stacks: (1) 1,200 Solidity v0.8.20-0.8.28 EVM DeFi smart contracts, (2) 1,200 Rust/Solana Anchor on-chain programs, (3) 1,200 Python/TypeScript Autonomous AI Agent, MCP, RAG & RLVR guardrail modules, (4) 1,200 Go/TypeScript/Python Cloud & FinTech API services, and (5) 1,200 Kubernetes, Terraform & CI/CD IaC configurations. Packaged in Apache Parquet, CSV, and JSONL with SHA-256 checksums. Zero overlap with public Hugging Face v1-v3 splits.

Format
ZIP
Downloads
0
Sample structureNot assessed
text
$249

Solidity & AI Agent Security Remediation Benchmark v4 (1,600-Row Core Split)

Updated Nov 11, 2026

Curated by Oxford Digital (oxforddigitals.com). 1,600 unseen code security & remediation pairs combining 800 Solidity v0.8.20-0.8.26 smart contract audit/patch records and 800 Python/TypeScript AI Agent, MCP tool server, RAG context-isolation, Text-to-SQL, and RLVR grader hardening records. Packaged in Apache Parquet, UTF-8 CSV, and JSONL with SHA-256 checksums. Built for LLM fine-tuning (SFT/DPO/GRPO/RLVR), AI coding copilots, and automated security benchmark evaluation. Cryptographically verified for zero overlap with public Hugging Face v1-v3 corpora.

Format
ZIP
Downloads
0
Sample structureNot assessed
text
$500

Master User-Agent and Device Detection Regex to Integer Mapping Matrix

3,152 Regex Keys · Updated Nov 11, 2026

A highly optimized, byte-sanitized reference index mapping raw global browser user-agent regex parameters to unique 4-byte integer tokens for 95% database storage reduction.

Records
3,151
Format
CSV
Downloads
0
Sample structure100/100
text
$39

Portugal Data API — Portuguese Open Data via MCP (Real Estate, Tenders, Porto Leads)

Updated Nov 11, 2026

Machine-readable dataset bundle backing the 'portugal-data-api' MCP server (Model Context Protocol, Streamable HTTP). Public endpoint: https://hermesportugal.duckdns.org/mcp-ptdata/ — 5 tools: get_real_estate_listings (daily casa.sapo.pt listings), get_procurement_tenders (base.gov.pt tender notices), get_porto_agency_leads, get_porto_agency_phones (outreach contacts for Porto estate agencies), get_api_status. Includes an OpenAPI 3.x spec of the underlying x402 Portugal Data API. Records are metadata/public-listings fields (no buyer personal data). Source: base.gov.pt (Portal de Contratos Públicos), casa.sapo.pt.

Records
5
Format
JSONL
Downloads
0
Sample structure62.5/100
text
$200

General Hausa Village Childhood Oral History - 37min Audio-Aligned EN-HA + 25-Term Zamfara Dialect & Royal Titles Lexicon

2026 · Updated Nov 1, 2026

A 37-minute natural, unscripted oral history in General Hausa (Zamfara dialect) with time-aligned transcriptions and English translations. This dataset includes a rare 25-term cultural lexicon (palace titles, traditional games). Ethically sourced with explicit speaker consent, it is ideal for training ASR, TTS, and translation models on low-resource languages.

Format
ZIP
Downloads
0
Sample structureNot assessed
text
$39

Portugal Public Procurement Tender Notices — base.gov.pt (250 records, Oct 2026)

Updated Oct 23, 2026

Structured dataset of public procurement tender notices (anuncios de procedimento) from base.gov.pt — Portugal's official public contracts portal. Fields per record: tender title, contracting entity (entidade), act type, procedure type (concurso publico, ajuste direto, etc.), base price in EUR, publication date, proposal deadline, and source URL. 250 most recent notices, publication dates 19-2026-10-22. Use cases: sales intelligence for suppliers bidding on public tenders, market monitoring of public spending, competitive analysis, RAG/agent workflows on EU public data. Public-sector information from the official portal (base.gov.pt); source credited in documentation. Metadata-only: no bidders' personal data, no emails/phones.

Records
250
Format
JSONL
Downloads
0
Sample structure100/100
text
$49

Classic English Novels, Computed: Sentiment Arcs, Genres and an Author Influence Graph (343 Novels)

Updated Oct 20, 2026

Computed features for 343 public-domain English novels (1660-1928): a 100-point sentiment arc per book with its shape, eight genres recovered from opening prose alone, book-to-book arc similarity, and a 77-author graph of documented influences with stylistic and conceptual similarity for every author pair. One CSV, 343 rows; column meanings in the schema. No book text. Genre labels from third-party reading lists, used only for validation, are left out.

Records
343
Format
CSV
Downloads
0
Sample structure100/100
text
$7

Labeled Customer Support Conversations

Synthetically generated multi-turn dialogues with intent, sentiment, and resolution labels across 25 support categories. Designed for training and evaluating customer support AI agents.

Records
10,000
Format
JSONL
Downloads
0
Sample structure100/100
text
Free

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
Sample structure100/100
text
Free

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
Sample structure100/100
text
Free

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
Sample structure100/100
text
Free

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
Sample structure100/100
text
Free

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
Sample structure100/100
text
Free

OpenMath GSM8K Masked Grade-School Math

Grade-school math word problems with masked and full reference solutions plus expected answers, designed to support synthetic solution generation.

Records
7,473
Format
PARQUET
Downloads
2
Sample structure100/100
text
Free

MiniPile Deduplicated Text Corpus

Compact subset of the deduplicated Pile corpus containing approximately 1 million documents, designed for efficient language-model pretraining and experimentation.

Records
1,010,500
Format
PARQUET
Downloads
3
Sample structure100/100
text
Free

PII Masking Dataset

Text samples annotated for personally identifiable information detection and masking, with paired source and target text, span labels, and token-level annotations across multiple languages.

Records
225,405
Format
PARQUET
Downloads
4
Sample structure100/100
text
Free

Stack Overflow Posts Markdown

before June 14, 2023

Questions and answers with scores, tags, view counts, creation and edit timestamps. Includes full post metadata for each record.

Records
58,329,355
Format
PARQUET
Downloads
4
Sample structure82.8/100
text
Free

Waveform-5000 Classification Benchmark

Synthetic 3-class waveform classification dataset with 40 numeric attributes, including 21 informative features and 19 noise dimensions. Originally from Breiman et al. 1984.

Records
5,000
Format
PARQUET
Downloads
6
Sample structure100/100
text
Free

Nemotron Content Safety Audio Dataset

English audio files covering 23 violation categories designed for evaluating multimodal content-safety guardrails. Extends NVIDIA's Aegis 2.0 benchmark into the audio modality with adversarial and safety-critical examples.

Records
1,928
Format
PARQUET
Downloads
4
Sample structure91.3/100
text
Free

Hindawi Arabic Books Section-Level NLP Dataset

Cleaned, section-level Arabic text from Hindawi.org books spanning literature, philosophy, history, and science, prepared for Arabic NLP training and research.

Records
52,830
Format
PARQUET
Downloads
2
Sample structure92.9/100
text
Free

MInDS-14 Multilingual Spoken Intent Detection

Spoken intent detection benchmark with e-banking intents across 14 language varieties, comprising audio recordings and transcriptions for speech understanding evaluation and multilingual ASR/NLU fine-tuning.

Records
16,336
Format
PARQUET
Downloads
3
Sample structure100/100
text
Free

AG News Topic Classification Benchmark

News articles classified into four topics: World, Sports, Business, and Science/Technology. Standard benchmark for evaluating text classification models and NLP systems.

Records
127,600
Format
PARQUET
Downloads
3
Sample structure100/100