# TypeSafe highlights MotherDuck prompt_jev() as ~50x faster, ~100x cheaper text classification

TypeSafe quoted MotherDuck’s claim that prompt_jev(), a SQL function powered by TypeSafe’s Jev model, made text classification about 50x faster at about 1% of the cost. MotherDuck reported 100k rows in 40 seconds for $0.50 versus 32 minutes and $37 for an LLM.

Language: en
Time zone: America/Los_Angeles
Current time (UTC, cached as of): 2026-11-10T06:49:00.000Z
Query date (America/Los_Angeles): 2026-11-09

HTML: https://didcodexreset.com/news/0f6d03ad25a84ae4ad9d172e.html

Content language: en
Localization state: sameLanguage

Source: TypeSafe AI · Published 9/21/2026, 18:28:20

TypeSafe AI quoted MotherDuck and emphasized Jev as **50x faster** and **100x cheaper** for adding natural-language intelligence in applications.

MotherDuck said **text classification** on its platform **just got ~50x faster at ~1% of the cost**. It described **prompt_jev()** as a SQL function powered by **Jev**, which it called TypeSafe’s new system one model, and reported that **100k rows** ran in **40s for $0.50** with frontier-LLM accuracy, compared with **32 min and $37** for “the LLM.” MotherDuck linked more detail at [motherduck.com/blog/motherduck-supports-jev/](https://motherduck.com/blog/motherduck-supports-jev/).

Tags: MotherDuck, Jev, TypeSafe, Text Classification

[View original post](https://x.com/typesafeai/status/2102206611024716181)