Compare language models with scored test suites.
Evaluates language models with scored test suites and presents results for model comparisons and CI checks.
When to use it
Use to create or run evaluation suites, inspect benchmark results, compare models, or integrate evaluations into CI.
Give it an evaluation request; it uses the CLI to return run details, per-sample results, or scorecards.
Requires the omniroute CLI; the skill includes a global installation command.
This skill
OmniRoute
Installs OmniRoute and starts evaluations
Stores evaluation suites and results