Home /Benchmarks, evals and calibration /jev-sec-bench
jev-sec-bench
Blind benchmarks for prompt injection and vulnerable-code detection.
From the readme
jev-sec-bench Blind security benchmarks for Jev, TypeSafe's System One model, built on jev-go. Jev does not generate text. It reads a state and returns typed judgments with calibrated probabilities, which is the shape a guardrail actually needs: a number your code can threshold, not a paragraph you have to parse. Two benchmarks, both blind, both on public …
Details
- Section
- Benchmarks, evals and calibration
- owner
- Gaurav-Gosain
- stars
- 2
- forks
- 0
- pushed
- 2026-09-16
- Language
- Go
- License
- MIT
What it is
- Form
- Dataset or benchmark
- Host agent
- Standalone
- Audience
- Researchers
Maturity
Docs
reports measured numbers