TMDB Movie & TV Metadata Scraper

Scrape rich metadata for movies and TV shows from The Movie Database (TMDB) — no API key required. Discovers titles from TMDB's public discover/browse pages and extracts full detail records including cast, directors, genres, keywords, ratings, runtime, status, certification, budget, revenue, and networks.

What you get

Each record in the output dataset contains:

Field	Description
tmdb_id	Numeric TMDB ID
title	Movie or TV show title
media_type	movie or tv
tmdb_url	Canonical TMDB page URL
original_title	Title in the original language
release_date	First release or air date (YYYY-MM-DD)
vote_average	Aggregate rating (0–10 scale)
vote_count	Number of votes
user_score_percent	User score percentage (0–100)
overview	Plot summary or show description
genres	Comma-separated genre names
runtime_minutes	Movie runtime in minutes (empty for TV: TMDB lists no series runtime)
original_language	Original language as an English name, for example English or Japanese
cast_top	Comma-separated top-billed cast names
directors	Movies only: comma-separated director names (TV pages list "Created by", not a director)
keywords	Comma-separated TMDB keyword tags
end_date	TV only: date the series last aired, empty while it is still running
status	Production status, for example Released, Returning Series, Ended
tagline	Marketing tagline, empty when TMDB has none
certification	Age rating, for example PG-13 or TV-MA, empty when none is listed
countries	Comma-separated countries of origin
budget_usd	Movies only: budget in USD. Empty when TMDB lists it as unknown (never 0)
revenue_usd	Movies only: worldwide revenue in USD. Empty when TMDB lists it as unknown (never 0)
networks	TV only: comma-separated networks
number_of_episodes	TV only: total episodes listed
poster_url	Poster image URL

Fields that TMDB does not publish for a title are returned empty rather than filled with a placeholder. original_title is only filled when it differs from title.

Why use this instead of the TMDB API?

Most TMDB scrapers on the Apify Store are thin wrappers around the TMDB REST API — they require you to register for and manage your own API key and stay within TMDB's per-account rate limits. This actor scrapes TMDB's public web pages directly, so:

• No API key registration or management
• No per-account rate limits to worry about
• Both movies and TV shows in one unified output schema
• Includes fields not always easily queryable via API (keyword tags, top cast, budget and revenue)

Inputs

Input	Type	Default	Description
maxItems	integer	15	Maximum number of records to return. Set to 0 for no limit. With both, the cap is shared evenly between movies and TV.
mediaType	string	both	Which media type to scrape: movie, tv, or both.
startPage	integer	1	Discover page to start from, 1 to 500 (each page has 20 titles).

Example use cases

• Media server catalogs: Build or enrich metadata catalogs for Plex, Jellyfin, or Kodi libraries without managing API credentials.
• Recommendation engines: Feed movie/TV metadata into ML pipelines — genres, keywords, cast, and ratings in one schema.
• Box office analysis: Compare budget and revenue across movies.
• Market research: Track ratings and popularity trends across the TMDB catalog over time.

How it works

1. Discover: Crawls paginated TMDB browse pages (/movie?language=en-US&page=N, /tv?...) — 20 titles per page, up to 500 pages per type.
2. Detail: For each title, fetches the detail page and extracts:
   • title, description, rating, genres, runtime, release and end dates, user score, directors, cast, keywords, status, networks, budget and revenue

Notes

• TMDB's discover pages order titles by popularity (most popular first). Use startPage to offset into the catalog.
• TMDB serves at most 500 pages of 20 titles per media type (10,000 titles). The run stops cleanly at that limit.
• Popularity, ratings and vote counts move constantly, so repeated runs return different values and ordering.
• The language=en-US parameter is appended to all requests to ensure English metadata in the output.