JevMade

Sign in
← Back to experiments

Benchmarks & research

Jev OOD Calibration

Puts Jev's confidence numbers to the test on a task that was generated after the model shipped.

Source screenshot of Jev OOD Calibration
SOURCE SCREENSHOTFull screenshot ↗

What it does

Nine hundred synthetic tickets and three public benchmarks later, the numbers hold up in public but drift out of distribution, and all raw responses are published for anyone to recheck.

How you can use it

Before writing any code, gather your own test examples with known correct answers. A developer can borrow this testing setup to check how Jev scores your actual tasks. Running these checks needs paid account credits to avoid tight request limits.

This research shows Jev can claim high certainty even when guessing on unfamiliar rules. Because its reported certainty numbers become unreliable on novel tasks, do not let the model approve important actions on its own.

Maker-reported (not independently measured by JevMade)

3,721 public-benchmark items + 900 synthetic tickets, 0 failed calls, about $0.06 of API calls

Primitives
choice, score, noul
Platform
Python
Added
Project created

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.