JevMade

Sign in
← Back to experiments

Benchmarks & research

open-jev (daseinlabs)

An open reimplementation trains a lightweight decision head for Choice, Score and Noul questions.

Source screenshot of open-jev (daseinlabs)
SOURCE SCREENSHOTFull screenshot ↗

What it does

Its published run reports 77 ms per example and roughly 15 seconds for eight training epochs.

How you can use it

A developer can connect this tool to a customer service app. The app reads a message about a wrong shoe size. It asks the tool several questions at once. The tool then picks the right department and decides if the customer wants a refund.

The software runs entirely on your own computer without an outside AI service. A developer must accept a license agreement before downloading the files. The computer needs fast equipment and plenty of memory. Without them, the program answers much more slowly.

Maker-reported (not independently measured by JevMade): 77 ms per example, head training 8 epochs in about 15 s

Primitives
choice, score, noul
Platform
Python
Added
Project created
GitHub stars
31 (snapshot, not live)

Keep this for later

Sign in to bookmark experiments, guides and videos, and keep notes only you can see.

Continue to sign in

We’ll bring you back to this listing.