OpenData Documentation

OpenData is a platform for discovering, accessing, and analyzing open datasets. Think of it like GitHub, but for data.

Quick Start

Find a dataset, query it over the REST API, and save the result as a view — in about five minutes.

Key Concepts

Provider: Organizations that publish data (e.g., BLS, Census, FRED). Each provider has a unique slug and can host multiple datasets.
Dataset: A collection of data from a provider. Datasets can be flat (single table) or hierarchical (multiple subdatasets).
Subdataset: For hierarchical datasets, individual data files. For example, BLS CPI-U has subdatasets for different data series.
View: A query-time transformation applied to a dataset. Views can reshape, filter, or join data without modifying the underlying storage. See the Views guide for details.

API Quick Start

All data is accessible via REST API. Here are the most common endpoints:

Request

# List all datasets
GET /v1/datasets

# Get dataset metadata
GET /v1/datasets/{provider}/{dataset}/meta

# Query dataset data
GET /v1/datasets/{provider}/{dataset}?limit=100

# Search datasets
GET /v1/search?q=population

For complete API documentation, see the API Reference.

Want the OpenAPI spec?

The full API is described by an OpenAPI spec at https://api.tryopendata.ai/openapi.json. Point it at a client generator, import it into Postman or Insomnia, or hand it to an AI agent. The API Reference page has copy and download buttons for it in the header.