Data source
Data from Twitch Developers, delivered clean.
2 datasets pulled from Twitch Developers's releases, checked field by field and shipped the way you want them — daily, weekly, or hourly, your call.
- 2 datasets
- 2 industrys
- Real rows on request
Interactive Media & Services
1Twitch Helix API – Live Streaming & Creator Economy Data
Interactive Home Entertainment
1TwitchTracker - Game & Channel Rankings
Pick a catch, see the rows.
Name any Twitch Developers dataset and we send real rows from it — not a screenshot of rows. 1,744 datasets. Pick your catch.
Get a sampleAPI, files, or your warehouse. Daily, weekly, or hourly.
Straight answers about Twitch Developers data
How much of Twitch does the data cover?
The catalog sizes the surface in the millions of concurrent live streams and channels, across the whole global platform, with each stream's language exposed as ISO 639-1 codes. Twitch publishes no aggregate totals anywhere in its data surface, so audience size arrives measured, row by row - which is exactly what a delivery provides.
Does the data include history, or only the live state?
The core tables describe the present: every current broadcast, its viewer count, its start timestamp. Clips and videos retain their own published history, and dated reporting ranges exist on the analytics side. A continuous viewership time series is built by scheduling deliveries on a fixed cadence - a decision made once, then automated.
Can a sample be cut to specific categories, languages or streamers?
Yes. Samples arrive scoped to the slice you name - particular game categories, languages, or a watchlist of broadcasters - before any commitment, so the join keys and field behavior are judged on the rows you would actually receive rather than a generic brochure extract.
Is the schema stable enough for production pipelines?
Field definitions were verified against live records during cataloging, and the dictionary ships unchanged across every delivery channel. Schema shifts get flagged rather than discovered, and the two known-deprecated stream fields are called out explicitly, so pipelines written against the dictionary do not break quietly.