DuraGym by Dura Labs Controlled Digital Intelligence

Open-weight model assurance

Is this model safe and secure enough for your use case?

DuraGym tests open-weight models against a catalog of benchmarks and your own acceptance criteria, shows why a model fails where it fails, and ends in a release and a report a risk owner can sign off on.

  1. 01Modelssource intelligence
  2. 02Benchmarkschoose assessments
  3. 03Evaluatetest your models
  4. 04Trainfix the behaviors
  5. 05Releasecut new versions

Sign in

Continue with an approved Google or GitHub account.

Continue with Google Continue with GitHub

Access is limited to approved accounts. To ask for access, write to Dura Labs.