← Catalog

Airbnb Data Download: 6-City Listings CSV

You need Airbnb listings data by city for a market study, a school project, or a script that pulls numbers on a schedule. Scraping Airbnb directly gets accounts blocked and breaks every time the page markup changes. Inside Airbnb publishes the numbers as CSV files instead, and this guide shows what is in those files, which cities are covered here, and how to query the full 90,169 rows without downloading anything.

Coverage: 6 cities, one snapshot per city

Each city was captured on one date, so every row is a picture of the market that day. Snapshot dates came from the source and are written here as-is:

City	Snapshot date	Listings	Median nightly price (USD)
Austin, TX	2026-06-22	11,295	217
Nashville, TN	2026-06-26	10,242	282
Denver, CO	2026-06-30	4,939	176
New York City, NY	2026-08-10	30,234	175
Las Vegas (Clark County), NV	2026-06-27	20,234	234
San Diego, CA	2026-06-27	13,225	323

Ninety thousand one hundred sixty-nine listings in total. New York City is the biggest slice at 30,234 rows. A city-level index with room-type mix per city ships with the sample files as cities-index.csv.

What is in listings-summary.csv

Every city uses the same 12 columns, so a query written for Austin runs unchanged against San Diego. The main ones:

id: Listing identifier, stable across snapshots on the source side.
name: The listing title as it appears on Airbnb.
neighbourhood: ZIP code area for the listings in this pack.
latitude, longitude: Approximate location. Inside Airbnb blurs coordinates by about 11 meters before publishing, which protects individual addresses while keeping the data usable for mapping.
room_type: Entire home/apt, Private room, Hotel room, or Shared room.
price: Nightly rate in USD, cleaned from the source text.
minimum_nights: Smallest stay the host accepts.
availability_365: Bookable days for the next 365 days from the snapshot date.
number_of_reviews, reviews_per_month, last_review: Review count, rate, and most recent review date.

No host identity in any column. Inside Airbnb strips host names and contact details from the summary file before publishing, and this pack keeps those fields out. That makes the file safe to hand to students or load into a shared database.

Download the 20-row sample CSV to see all 12 columns with real Austin rows, or read the machine-readable sample.json.

Querying without scraping

The full pack is wired to a read-only MCP server, so an agent or script can filter and paginate the complete 90,169-row files over the protocol. Two tools do the work:

query_dataset(slug, {where, columns, limit, offset}) filters and paginates the full released files, 100 rows per call, with =, contains, gt, and lt operators.
get_stats(slug, {column}) returns unique counts, numeric min/max/mean, and top-N values.

Connect by running the server from the repo:

npx --allow-git=all -y github:jayjex/dataset-mcp

An example call for one city:

query_dataset("airbnb-six-cities", {where: [{column: "neighbourhood", op: "=", value: "78701"}], limit: 5})

Endpoint docs with more worked examples: For agents. The catalog entry for this dataset lives in catalog.json.

License and attribution

All listing data comes from Inside Airbnb and is licensed CC BY 4.0. If you republish the free samples, credit the source as Inside Airbnb, CC BY 4.0, plus Data Vault (jayjex.github.io/data-vault). The full pack carries the same upstream license.

Download

Download free sample (CSV) Get full data

Full pack: 90,169 rows across 6 cities, per-city CSVs, data dictionary, methodology, source checksums. Checkout and download run through Getly. Want to try before you commit? Start from the free Airbnb dataset sample.