GoKawiilResearcher scrapes 4.5 billion TikTok videos via reverse-engineered Android API, publishes dataset on Hugging Face
A developer detailed how to bypass TikTok's public web endpoints and instead access the private HTTP+JSON API used by its Android app, which returns far more data at higher speed. Using this method, the researcher collected 3.23 billion creator profiles, 5.94 billion videos and 2.8 billion comments in three weeks, and has released 4.5 billion of those videos as a free dataset on Hugging Face with captions, engagement metrics, sound data, country and posting time.
GoKawiil's interpretation of the reporting above, not reported fact.
This exposes how fragile TikTok's data protections are against anyone willing to mimic device credentials, request signatures and TLS fingerprints, undermining assumptions about scraping limits on major platforms. The public release of billions of creator and video records raises fresh privacy and platform-security concerns, since it could be used for anything from research to targeted advertising or harassment at massive scale.
- A reverse-engineered Android API let researchers pull vastly more TikTok data than public endpoints allow.
- The resulting dataset of 4.5 billion videos, including engagement stats and metadata, is now public on Hugging Face.
- The scraping method required spoofing device credentials, signatures, regional hosts and TLS fingerprints simultaneously, exposing a security gap.
Source: tiktok-api.seeksocial.io, 2026-09-03
Published there as: “4.5B Posts Scraped from TikTok”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.