What it does
The primary arm is the open Kev-9B, which anyone can run, with hosted Jev as a second arm — the arrangement TypeSafe's paused signups forced.
Benchmarks & research
jev-guardbench
Asks whether a System One model can do the job of an LLM judge in an agent guardrail, with the hypotheses and decision rules fixed before any run.
Bookmark: jev-guardbench
Keep this in your collection.
Leave a noteWhat would you try with this?
: jev-guardbench
Only you can see your notes.
Screenshot unavailable. Open the experiment ↗
How you can use it
If your app uses AI to check messages before sending them, ask a developer to compare the checks with this project. Begin with its small trial run. Look at both mistakes and waiting time. A faster checker is not useful if it misses more problems.