Rhetra Get in touch
Platform · Pipeline

A pipeline, running.

Fourteen Reddit searches, one per model name. Stacked, deduped, split, then classified by Haiku and written straight into the ontology. Nothing is staged: this is one run, from the click to the rows.

Wall clock
178.1s
LLM calls
779
Tokens in
689,591
Cost
$0.7368
Datasets
25
Entities
189

The clip runs 22 seconds. The first four and the last ten are real time, so you see the run actually start and actually land; the 170 seconds of grinding in between are compressed hard. The last thing to appear is both_labelled, 184 rows, which is the point: the run ends as queryable objects, not as a log line.

See the rest of the platform Worked examples