DATAtokeniz
New✦ Any source in, training data out

Easily make ML-ready datasets

Bring your data, tell DATAtokeniz how to handle it, and its deterministic pipeline compiles it — with exact provenance on every row.

PDF → graded eval setno key needed

Star on GitHubStart for free

Transformer

Input → guided handling → JSONL output

Input

Source material

PDF, DOCX, slides, CSV, JSON, images, audio, video or a link. Processed in the request and never stored. Audio and video are transcribed into timestamped segments; a public Kaggle dataset is downloaded and every row cites its file.

Declare source license— optional

Demo mode · nothing is saved. Sign in to raise the limits and keep a run history.

Transformer

Tell DATAtokeniz how to handle it

Guide

Add a source on the left, then tell me exactly how to turn it into a dataset.

  • extract
  • segment
  • enrich
  • quality
  • compile

Extracted only — the guide never writes dataset rows.

Output

ML-ready dataset

Your dataset appears here after a run — preview, quality score and one-click download.