Tech News
← Home  ·  All topics

Gliner

2 GoKawiil briefs on this topic

Analysis of Reddit knife forums finds concentrated brand-boosting accounts

A data analysis by blogger Peter Vijeh examined comments across six Reddit knife subreddits using a fine-tuned named-entity recognition model to track brand mentions. He found that for one chef's-knife brand, 31% of its mentions in 'what should I buy' threads came from just 5% of accounts, a rate four times higher than random chance would predict, prompting him to review those accounts' full posting histories.

Developer fine-tunes open-source GLiNER model using Gemini-labeled data for $9

A hobbyist building a knife-tracking scraper for Reddit comments replaced expensive per-comment Gemini 3.1 Pro API calls with a fine-tuned version of the open-source GLiNER model. He had Gemini label 4,290 comments once, then used that labeled data to train GLiNER large v2.5 to extract knife brands, models and steel types, spending about $9 total on the labeling step.